path: root/doc/filters.texi (plain)
blob: 8e5e21f9eda97eaed3942015b79f4bf1c7397022

1	@chapter Filtering Introduction
2	@c man begin FILTERING INTRODUCTION
3
4	Filtering in FFmpeg is enabled through the libavfilter library.
5
6	In libavfilter, a filter can have multiple inputs and multiple
7	outputs.
8	To illustrate the sorts of things that are possible, we consider the
9	following filtergraph.
10
11	@verbatim
12	[main]
13	input --> split ---------------------> overlay --> output
14	| ^
15	|[tmp] [flip]|
16	+-----> crop --> vflip -------+
17	@end verbatim
18
19	This filtergraph splits the input stream in two streams, then sends one
20	stream through the crop filter and the vflip filter, before merging it
21	back with the other stream by overlaying it on top. You can use the
22	following command to achieve this:
23
24	@example
25	ffmpeg -i INPUT -vf "split [main][tmp]; [tmp] crop=iw:ih/2:0:0, vflip [flip]; [main][flip] overlay=0:H/2" OUTPUT
26	@end example
27
28	The result will be that the top half of the video is mirrored
29	onto the bottom half of the output video.
30
31	Filters in the same linear chain are separated by commas, and distinct
32	linear chains of filters are separated by semicolons. In our example,
33	@var{crop,vflip} are in one linear chain, @var{split} and
34	@var{overlay} are separately in another. The points where the linear
35	chains join are labelled by names enclosed in square brackets. In the
36	example, the split filter generates two outputs that are associated to
37	the labels @var{[main]} and @var{[tmp]}.
38
39	The stream sent to the second output of @var{split}, labelled as
40	@var{[tmp]}, is processed through the @var{crop} filter, which crops
41	away the lower half part of the video, and then vertically flipped. The
42	@var{overlay} filter takes in input the first unchanged output of the
43	split filter (which was labelled as @var{[main]}), and overlay on its
44	lower half the output generated by the @var{crop,vflip} filterchain.
45
46	Some filters take in input a list of parameters: they are specified
47	after the filter name and an equal sign, and are separated from each other
48	by a colon.
49
50	There exist so-called @var{source filters} that do not have an
51	audio/video input, and @var{sink filters} that will not have audio/video
52	output.
53
54	@c man end FILTERING INTRODUCTION
55
56	@chapter graph2dot
57	@c man begin GRAPH2DOT
58
59	The @file{graph2dot} program included in the FFmpeg @file{tools}
60	directory can be used to parse a filtergraph description and issue a
61	corresponding textual representation in the dot language.
62
63	Invoke the command:
64	@example
65	graph2dot -h
66	@end example
67
68	to see how to use @file{graph2dot}.
69
70	You can then pass the dot description to the @file{dot} program (from
71	the graphviz suite of programs) and obtain a graphical representation
72	of the filtergraph.
73
74	For example the sequence of commands:
75	@example
76	echo @var{GRAPH_DESCRIPTION} | \
77	tools/graph2dot -o graph.tmp && \
78	dot -Tpng graph.tmp -o graph.png && \
79	display graph.png
80	@end example
81
82	can be used to create and display an image representing the graph
83	described by the @var{GRAPH_DESCRIPTION} string. Note that this string must be
84	a complete self-contained graph, with its inputs and outputs explicitly defined.
85	For example if your command line is of the form:
86	@example
87	ffmpeg -i infile -vf scale=640:360 outfile
88	@end example
89	your @var{GRAPH_DESCRIPTION} string will need to be of the form:
90	@example
91	nullsrc,scale=640:360,nullsink
92	@end example
93	you may also need to set the @var{nullsrc} parameters and add a @var{format}
94	filter in order to simulate a specific input file.
95
96	@c man end GRAPH2DOT
97
98	@chapter Filtergraph description
99	@c man begin FILTERGRAPH DESCRIPTION
100
101	A filtergraph is a directed graph of connected filters. It can contain
102	cycles, and there can be multiple links between a pair of
103	filters. Each link has one input pad on one side connecting it to one
104	filter from which it takes its input, and one output pad on the other
105	side connecting it to one filter accepting its output.
106
107	Each filter in a filtergraph is an instance of a filter class
108	registered in the application, which defines the features and the
109	number of input and output pads of the filter.
110
111	A filter with no input pads is called a "source", and a filter with no
112	output pads is called a "sink".
113
114	@anchor{Filtergraph syntax}
115	@section Filtergraph syntax
116
117	A filtergraph has a textual representation, which is recognized by the
118	@option{-filter}/@option{-vf}/@option{-af} and
119	@option{-filter_complex} options in @command{ffmpeg} and
120	@option{-vf}/@option{-af} in @command{ffplay}, and by the
121	@code{avfilter_graph_parse_ptr()} function defined in
122	@file{libavfilter/avfilter.h}.
123
124	A filterchain consists of a sequence of connected filters, each one
125	connected to the previous one in the sequence. A filterchain is
126	represented by a list of ","-separated filter descriptions.
127
128	A filtergraph consists of a sequence of filterchains. A sequence of
129	filterchains is represented by a list of ";"-separated filterchain
130	descriptions.
131
132	A filter is represented by a string of the form:
133	[@var{in_link_1}]...[@var{in_link_N}]@var{filter_name}=@var{arguments}[@var{out_link_1}]...[@var{out_link_M}]
134
135	@var{filter_name} is the name of the filter class of which the
136	described filter is an instance of, and has to be the name of one of
137	the filter classes registered in the program.
138	The name of the filter class is optionally followed by a string
139	"=@var{arguments}".
140
141	@var{arguments} is a string which contains the parameters used to
142	initialize the filter instance. It may have one of two forms:
143	@itemize
144
145	@item
146	A ':'-separated list of @var{key=value} pairs.
147
148	@item
149	A ':'-separated list of @var{value}. In this case, the keys are assumed to be
150	the option names in the order they are declared. E.g. the @code{fade} filter
151	declares three options in this order -- @option{type}, @option{start_frame} and
152	@option{nb_frames}. Then the parameter list @var{in:0:30} means that the value
153	@var{in} is assigned to the option @option{type}, @var{0} to
154	@option{start_frame} and @var{30} to @option{nb_frames}.
155
156	@item
157	A ':'-separated list of mixed direct @var{value} and long @var{key=value}
158	pairs. The direct @var{value} must precede the @var{key=value} pairs, and
159	follow the same constraints order of the previous point. The following
160	@var{key=value} pairs can be set in any preferred order.
161
162	@end itemize
163
164	If the option value itself is a list of items (e.g. the @code{format} filter
165	takes a list of pixel formats), the items in the list are usually separated by
166	@samp{|}.
167
168	The list of arguments can be quoted using the character @samp{'} as initial
169	and ending mark, and the character @samp{\} for escaping the characters
170	within the quoted text; otherwise the argument string is considered
171	terminated when the next special character (belonging to the set
172	@samp{[]=;,}) is encountered.
173
174	The name and arguments of the filter are optionally preceded and
175	followed by a list of link labels.
176	A link label allows one to name a link and associate it to a filter output
177	or input pad. The preceding labels @var{in_link_1}
178	... @var{in_link_N}, are associated to the filter input pads,
179	the following labels @var{out_link_1} ... @var{out_link_M}, are
180	associated to the output pads.
181
182	When two link labels with the same name are found in the
183	filtergraph, a link between the corresponding input and output pad is
184	created.
185
186	If an output pad is not labelled, it is linked by default to the first
187	unlabelled input pad of the next filter in the filterchain.
188	For example in the filterchain
189	@example
190	nullsrc, split[L1], [L2]overlay, nullsink
191	@end example
192	the split filter instance has two output pads, and the overlay filter
193	instance two input pads. The first output pad of split is labelled
194	"L1", the first input pad of overlay is labelled "L2", and the second
195	output pad of split is linked to the second input pad of overlay,
196	which are both unlabelled.
197
198	In a filter description, if the input label of the first filter is not
199	specified, "in" is assumed; if the output label of the last filter is not
200	specified, "out" is assumed.
201
202	In a complete filterchain all the unlabelled filter input and output
203	pads must be connected. A filtergraph is considered valid if all the
204	filter input and output pads of all the filterchains are connected.
205
206	Libavfilter will automatically insert @ref{scale} filters where format
207	conversion is required. It is possible to specify swscale flags
208	for those automatically inserted scalers by prepending
209	@code{sws_flags=@var{flags};}
210	to the filtergraph description.
211
212	Here is a BNF description of the filtergraph syntax:
213	@example
214	@var{NAME} ::= sequence of alphanumeric characters and '_'
215	@var{LINKLABEL} ::= "[" @var{NAME} "]"
216	@var{LINKLABELS} ::= @var{LINKLABEL} [@var{LINKLABELS}]
217	@var{FILTER_ARGUMENTS} ::= sequence of chars (possibly quoted)
218	@var{FILTER} ::= [@var{LINKLABELS}] @var{NAME} ["=" @var{FILTER_ARGUMENTS}] [@var{LINKLABELS}]
219	@var{FILTERCHAIN} ::= @var{FILTER} [,@var{FILTERCHAIN}]
220	@var{FILTERGRAPH} ::= [sws_flags=@var{flags};] @var{FILTERCHAIN} [;@var{FILTERGRAPH}]
221	@end example
222
223	@section Notes on filtergraph escaping
224
225	Filtergraph description composition entails several levels of
226	escaping. See @ref{quoting_and_escaping,,the "Quoting and escaping"
227	section in the ffmpeg-utils(1) manual,ffmpeg-utils} for more
228	information about the employed escaping procedure.
229
230	A first level escaping affects the content of each filter option
231	value, which may contain the special character @code{:} used to
232	separate values, or one of the escaping characters @code{\'}.
233
234	A second level escaping affects the whole filter description, which
235	may contain the escaping characters @code{\'} or the special
236	characters @code{[],;} used by the filtergraph description.
237
238	Finally, when you specify a filtergraph on a shell commandline, you
239	need to perform a third level escaping for the shell special
240	characters contained within it.
241
242	For example, consider the following string to be embedded in
243	the @ref{drawtext} filter description @option{text} value:
244	@example
245	this is a 'string': may contain one, or more, special characters
246	@end example
247
248	This string contains the @code{'} special escaping character, and the
249	@code{:} special character, so it needs to be escaped in this way:
250	@example
251	text=this is a \'string\'\: may contain one, or more, special characters
252	@end example
253
254	A second level of escaping is required when embedding the filter
255	description in a filtergraph description, in order to escape all the
256	filtergraph special characters. Thus the example above becomes:
257	@example
258	drawtext=text=this is a \\\'string\\\'\\: may contain one\, or more\, special characters
259	@end example
260	(note that in addition to the @code{\'} escaping special characters,
261	also @code{,} needs to be escaped).
262
263	Finally an additional level of escaping is needed when writing the
264	filtergraph description in a shell command, which depends on the
265	escaping rules of the adopted shell. For example, assuming that
266	@code{\} is special and needs to be escaped with another @code{\}, the
267	previous string will finally result in:
268	@example
269	-vf "drawtext=text=this is a \\\\\\'string\\\\\\'\\\\: may contain one\\, or more\\, special characters"
270	@end example
271
272	@chapter Timeline editing
273
274	Some filters support a generic @option{enable} option. For the filters
275	supporting timeline editing, this option can be set to an expression which is
276	evaluated before sending a frame to the filter. If the evaluation is non-zero,
277	the filter will be enabled, otherwise the frame will be sent unchanged to the
278	next filter in the filtergraph.
279
280	The expression accepts the following values:
281	@table @samp
282	@item t
283	timestamp expressed in seconds, NAN if the input timestamp is unknown
284
285	@item n
286	sequential number of the input frame, starting from 0
287
288	@item pos
289	the position in the file of the input frame, NAN if unknown
290
291	@item w
292	@item h
293	width and height of the input frame if video
294	@end table
295
296	Additionally, these filters support an @option{enable} command that can be used
297	to re-define the expression.
298
299	Like any other filtering option, the @option{enable} option follows the same
300	rules.
301
302	For example, to enable a blur filter (@ref{smartblur}) from 10 seconds to 3
303	minutes, and a @ref{curves} filter starting at 3 seconds:
304	@example
305	smartblur = enable='between(t,10,3*60)',
306	curves = enable='gte(t,3)' : preset=cross_process
307	@end example
308
309	See @code{ffmpeg -filters} to view which filters have timeline support.
310
311	@c man end FILTERGRAPH DESCRIPTION
312
313	@chapter Audio Filters
314	@c man begin AUDIO FILTERS
315
316	When you configure your FFmpeg build, you can disable any of the
317	existing filters using @code{--disable-filters}.
318	The configure output will show the audio filters included in your
319	build.
320
321	Below is a description of the currently available audio filters.
322
323	@section acompressor
324
325	A compressor is mainly used to reduce the dynamic range of a signal.
326	Especially modern music is mostly compressed at a high ratio to
327	improve the overall loudness. It's done to get the highest attention
328	of a listener, "fatten" the sound and bring more "power" to the track.
329	If a signal is compressed too much it may sound dull or "dead"
330	afterwards or it may start to "pump" (which could be a powerful effect
331	but can also destroy a track completely).
332	The right compression is the key to reach a professional sound and is
333	the high art of mixing and mastering. Because of its complex settings
334	it may take a long time to get the right feeling for this kind of effect.
335
336	Compression is done by detecting the volume above a chosen level
337	@code{threshold} and dividing it by the factor set with @code{ratio}.
338	So if you set the threshold to -12dB and your signal reaches -6dB a ratio
339	of 2:1 will result in a signal at -9dB. Because an exact manipulation of
340	the signal would cause distortion of the waveform the reduction can be
341	levelled over the time. This is done by setting "Attack" and "Release".
342	@code{attack} determines how long the signal has to rise above the threshold
343	before any reduction will occur and @code{release} sets the time the signal
344	has to fall below the threshold to reduce the reduction again. Shorter signals
345	than the chosen attack time will be left untouched.
346	The overall reduction of the signal can be made up afterwards with the
347	@code{makeup} setting. So compressing the peaks of a signal about 6dB and
348	raising the makeup to this level results in a signal twice as loud than the
349	source. To gain a softer entry in the compression the @code{knee} flattens the
350	hard edge at the threshold in the range of the chosen decibels.
351
352	The filter accepts the following options:
353
354	@table @option
355	@item level_in
356	Set input gain. Default is 1. Range is between 0.015625 and 64.
357
358	@item threshold
359	If a signal of second stream rises above this level it will affect the gain
360	reduction of the first stream.
361	By default it is 0.125. Range is between 0.00097563 and 1.
362
363	@item ratio
364	Set a ratio by which the signal is reduced. 1:2 means that if the level
365	rose 4dB above the threshold, it will be only 2dB above after the reduction.
366	Default is 2. Range is between 1 and 20.
367
368	@item attack
369	Amount of milliseconds the signal has to rise above the threshold before gain
370	reduction starts. Default is 20. Range is between 0.01 and 2000.
371
372	@item release
373	Amount of milliseconds the signal has to fall below the threshold before
374	reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
375
376	@item makeup
377	Set the amount by how much signal will be amplified after processing.
378	Default is 2. Range is from 1 and 64.
379
380	@item knee
381	Curve the sharp knee around the threshold to enter gain reduction more softly.
382	Default is 2.82843. Range is between 1 and 8.
383
384	@item link
385	Choose if the @code{average} level between all channels of input stream
386	or the louder(@code{maximum}) channel of input stream affects the
387	reduction. Default is @code{average}.
388
389	@item detection
390	Should the exact signal be taken in case of @code{peak} or an RMS one in case
391	of @code{rms}. Default is @code{rms} which is mostly smoother.
392
393	@item mix
394	How much to use compressed signal in output. Default is 1.
395	Range is between 0 and 1.
396	@end table
397
398	@section acrossfade
399
400	Apply cross fade from one input audio stream to another input audio stream.
401	The cross fade is applied for specified duration near the end of first stream.
402
403	The filter accepts the following options:
404
405	@table @option
406	@item nb_samples, ns
407	Specify the number of samples for which the cross fade effect has to last.
408	At the end of the cross fade effect the first input audio will be completely
409	silent. Default is 44100.
410
411	@item duration, d
412	Specify the duration of the cross fade effect. See
413	@ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
414	for the accepted syntax.
415	By default the duration is determined by @var{nb_samples}.
416	If set this option is used instead of @var{nb_samples}.
417
418	@item overlap, o
419	Should first stream end overlap with second stream start. Default is enabled.
420
421	@item curve1
422	Set curve for cross fade transition for first stream.
423
424	@item curve2
425	Set curve for cross fade transition for second stream.
426
427	For description of available curve types see @ref{afade} filter description.
428	@end table
429
430	@subsection Examples
431
432	@itemize
433	@item
434	Cross fade from one input to another:
435	@example
436	ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:c1=exp:c2=exp output.flac
437	@end example
438
439	@item
440	Cross fade from one input to another but without overlapping:
441	@example
442	ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:o=0:c1=exp:c2=exp output.flac
443	@end example
444	@end itemize
445
446	@section acrusher
447
448	Reduce audio bit resolution.
449
450	This filter is bit crusher with enhanced functionality. A bit crusher
451	is used to audibly reduce number of bits an audio signal is sampled
452	with. This doesn't change the bit depth at all, it just produces the
453	effect. Material reduced in bit depth sounds more harsh and "digital".
454	This filter is able to even round to continuous values instead of discrete
455	bit depths.
456	Additionally it has a D/C offset which results in different crushing of
457	the lower and the upper half of the signal.
458	An Anti-Aliasing setting is able to produce "softer" crushing sounds.
459
460	Another feature of this filter is the logarithmic mode.
461	This setting switches from linear distances between bits to logarithmic ones.
462	The result is a much more "natural" sounding crusher which doesn't gate low
463	signals for example. The human ear has a logarithmic perception, too
464	so this kind of crushing is much more pleasant.
465	Logarithmic crushing is also able to get anti-aliased.
466
467	The filter accepts the following options:
468
469	@table @option
470	@item level_in
471	Set level in.
472
473	@item level_out
474	Set level out.
475
476	@item bits
477	Set bit reduction.
478
479	@item mix
480	Set mixing amount.
481
482	@item mode
483	Can be linear: @code{lin} or logarithmic: @code{log}.
484
485	@item dc
486	Set DC.
487
488	@item aa
489	Set anti-aliasing.
490
491	@item samples
492	Set sample reduction.
493
494	@item lfo
495	Enable LFO. By default disabled.
496
497	@item lforange
498	Set LFO range.
499
500	@item lforate
501	Set LFO rate.
502	@end table
503
504	@section adelay
505
506	Delay one or more audio channels.
507
508	Samples in delayed channel are filled with silence.
509
510	The filter accepts the following option:
511
512	@table @option
513	@item delays
514	Set list of delays in milliseconds for each channel separated by '|'.
515	At least one delay greater than 0 should be provided.
516	Unused delays will be silently ignored. If number of given delays is
517	smaller than number of channels all remaining channels will not be delayed.
518	If you want to delay exact number of samples, append 'S' to number.
519	@end table
520
521	@subsection Examples
522
523	@itemize
524	@item
525	Delay first channel by 1.5 seconds, the third channel by 0.5 seconds and leave
526	the second channel (and any other channels that may be present) unchanged.
527	@example
528	adelay=1500|0|500
529	@end example
530
531	@item
532	Delay second channel by 500 samples, the third channel by 700 samples and leave
533	the first channel (and any other channels that may be present) unchanged.
534	@example
535	adelay=0|500S|700S
536	@end example
537	@end itemize
538
539	@section aecho
540
541	Apply echoing to the input audio.
542
543	Echoes are reflected sound and can occur naturally amongst mountains
544	(and sometimes large buildings) when talking or shouting; digital echo
545	effects emulate this behaviour and are often used to help fill out the
546	sound of a single instrument or vocal. The time difference between the
547	original signal and the reflection is the @code{delay}, and the
548	loudness of the reflected signal is the @code{decay}.
549	Multiple echoes can have different delays and decays.
550
551	A description of the accepted parameters follows.
552
553	@table @option
554	@item in_gain
555	Set input gain of reflected signal. Default is @code{0.6}.
556
557	@item out_gain
558	Set output gain of reflected signal. Default is @code{0.3}.
559
560	@item delays
561	Set list of time intervals in milliseconds between original signal and reflections
562	separated by '|'. Allowed range for each @code{delay} is @code{(0 - 90000.0]}.
563	Default is @code{1000}.
564
565	@item decays
566	Set list of loudnesses of reflected signals separated by '|'.
567	Allowed range for each @code{decay} is @code{(0 - 1.0]}.
568	Default is @code{0.5}.
569	@end table
570
571	@subsection Examples
572
573	@itemize
574	@item
575	Make it sound as if there are twice as many instruments as are actually playing:
576	@example
577	aecho=0.8:0.88:60:0.4
578	@end example
579
580	@item
581	If delay is very short, then it sound like a (metallic) robot playing music:
582	@example
583	aecho=0.8:0.88:6:0.4
584	@end example
585
586	@item
587	A longer delay will sound like an open air concert in the mountains:
588	@example
589	aecho=0.8:0.9:1000:0.3
590	@end example
591
592	@item
593	Same as above but with one more mountain:
594	@example
595	aecho=0.8:0.9:1000|1800:0.3|0.25
596	@end example
597	@end itemize
598
599	@section aemphasis
600	Audio emphasis filter creates or restores material directly taken from LPs or
601	emphased CDs with different filter curves. E.g. to store music on vinyl the
602	signal has to be altered by a filter first to even out the disadvantages of
603	this recording medium.
604	Once the material is played back the inverse filter has to be applied to
605	restore the distortion of the frequency response.
606
607	The filter accepts the following options:
608
609	@table @option
610	@item level_in
611	Set input gain.
612
613	@item level_out
614	Set output gain.
615
616	@item mode
617	Set filter mode. For restoring material use @code{reproduction} mode, otherwise
618	use @code{production} mode. Default is @code{reproduction} mode.
619
620	@item type
621	Set filter type. Selects medium. Can be one of the following:
622
623	@table @option
624	@item col
625	select Columbia.
626	@item emi
627	select EMI.
628	@item bsi
629	select BSI (78RPM).
630	@item riaa
631	select RIAA.
632	@item cd
633	select Compact Disc (CD).
634	@item 50fm
635	select 50µs (FM).
636	@item 75fm
637	select 75µs (FM).
638	@item 50kf
639	select 50µs (FM-KF).
640	@item 75kf
641	select 75µs (FM-KF).
642	@end table
643	@end table
644
645	@section aeval
646
647	Modify an audio signal according to the specified expressions.
648
649	This filter accepts one or more expressions (one for each channel),
650	which are evaluated and used to modify a corresponding audio signal.
651
652	It accepts the following parameters:
653
654	@table @option
655	@item exprs
656	Set the '|'-separated expressions list for each separate channel. If
657	the number of input channels is greater than the number of
658	expressions, the last specified expression is used for the remaining
659	output channels.
660
661	@item channel_layout, c
662	Set output channel layout. If not specified, the channel layout is
663	specified by the number of expressions. If set to @samp{same}, it will
664	use by default the same input channel layout.
665	@end table
666
667	Each expression in @var{exprs} can contain the following constants and functions:
668
669	@table @option
670	@item ch
671	channel number of the current expression
672
673	@item n
674	number of the evaluated sample, starting from 0
675
676	@item s
677	sample rate
678
679	@item t
680	time of the evaluated sample expressed in seconds
681
682	@item nb_in_channels
683	@item nb_out_channels
684	input and output number of channels
685
686	@item val(CH)
687	the value of input channel with number @var{CH}
688	@end table
689
690	Note: this filter is slow. For faster processing you should use a
691	dedicated filter.
692
693	@subsection Examples
694
695	@itemize
696	@item
697	Half volume:
698	@example
699	aeval=val(ch)/2:c=same
700	@end example
701
702	@item
703	Invert phase of the second channel:
704	@example
705	aeval=val(0)|-val(1)
706	@end example
707	@end itemize
708
709	@anchor{afade}
710	@section afade
711
712	Apply fade-in/out effect to input audio.
713
714	A description of the accepted parameters follows.
715
716	@table @option
717	@item type, t
718	Specify the effect type, can be either @code{in} for fade-in, or
719	@code{out} for a fade-out effect. Default is @code{in}.
720
721	@item start_sample, ss
722	Specify the number of the start sample for starting to apply the fade
723	effect. Default is 0.
724
725	@item nb_samples, ns
726	Specify the number of samples for which the fade effect has to last. At
727	the end of the fade-in effect the output audio will have the same
728	volume as the input audio, at the end of the fade-out transition
729	the output audio will be silence. Default is 44100.
730
731	@item start_time, st
732	Specify the start time of the fade effect. Default is 0.
733	The value must be specified as a time duration; see
734	@ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
735	for the accepted syntax.
736	If set this option is used instead of @var{start_sample}.
737
738	@item duration, d
739	Specify the duration of the fade effect. See
740	@ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
741	for the accepted syntax.
742	At the end of the fade-in effect the output audio will have the same
743	volume as the input audio, at the end of the fade-out transition
744	the output audio will be silence.
745	By default the duration is determined by @var{nb_samples}.
746	If set this option is used instead of @var{nb_samples}.
747
748	@item curve
749	Set curve for fade transition.
750
751	It accepts the following values:
752	@table @option
753	@item tri
754	select triangular, linear slope (default)
755	@item qsin
756	select quarter of sine wave
757	@item hsin
758	select half of sine wave
759	@item esin
760	select exponential sine wave
761	@item log
762	select logarithmic
763	@item ipar
764	select inverted parabola
765	@item qua
766	select quadratic
767	@item cub
768	select cubic
769	@item squ
770	select square root
771	@item cbr
772	select cubic root
773	@item par
774	select parabola
775	@item exp
776	select exponential
777	@item iqsin
778	select inverted quarter of sine wave
779	@item ihsin
780	select inverted half of sine wave
781	@item dese
782	select double-exponential seat
783	@item desi
784	select double-exponential sigmoid
785	@end table
786	@end table
787
788	@subsection Examples
789
790	@itemize
791	@item
792	Fade in first 15 seconds of audio:
793	@example
794	afade=t=in:ss=0:d=15
795	@end example
796
797	@item
798	Fade out last 25 seconds of a 900 seconds audio:
799	@example
800	afade=t=out:st=875:d=25
801	@end example
802	@end itemize
803
804	@section afftfilt
805	Apply arbitrary expressions to samples in frequency domain.
806
807	@table @option
808	@item real
809	Set frequency domain real expression for each separate channel separated
810	by '|'. Default is "1".
811	If the number of input channels is greater than the number of
812	expressions, the last specified expression is used for the remaining
813	output channels.
814
815	@item imag
816	Set frequency domain imaginary expression for each separate channel
817	separated by '|'. If not set, @var{real} option is used.
818
819	Each expression in @var{real} and @var{imag} can contain the following
820	constants:
821
822	@table @option
823	@item sr
824	sample rate
825
826	@item b
827	current frequency bin number
828
829	@item nb
830	number of available bins
831
832	@item ch
833	channel number of the current expression
834
835	@item chs
836	number of channels
837
838	@item pts
839	current frame pts
840	@end table
841
842	@item win_size
843	Set window size.
844
845	It accepts the following values:
846	@table @samp
847	@item w16
848	@item w32
849	@item w64
850	@item w128
851	@item w256
852	@item w512
853	@item w1024
854	@item w2048
855	@item w4096
856	@item w8192
857	@item w16384
858	@item w32768
859	@item w65536
860	@end table
861	Default is @code{w4096}
862
863	@item win_func
864	Set window function. Default is @code{hann}.
865
866	@item overlap
867	Set window overlap. If set to 1, the recommended overlap for selected
868	window function will be picked. Default is @code{0.75}.
869	@end table
870
871	@subsection Examples
872
873	@itemize
874	@item
875	Leave almost only low frequencies in audio:
876	@example
877	afftfilt="1-clip((b/nb)*b,0,1)"
878	@end example
879	@end itemize
880
881	@anchor{aformat}
882	@section aformat
883
884	Set output format constraints for the input audio. The framework will
885	negotiate the most appropriate format to minimize conversions.
886
887	It accepts the following parameters:
888	@table @option
889
890	@item sample_fmts
891	A '|'-separated list of requested sample formats.
892
893	@item sample_rates
894	A '|'-separated list of requested sample rates.
895
896	@item channel_layouts
897	A '|'-separated list of requested channel layouts.
898
899	See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
900	for the required syntax.
901	@end table
902
903	If a parameter is omitted, all values are allowed.
904
905	Force the output to either unsigned 8-bit or signed 16-bit stereo
906	@example
907	aformat=sample_fmts=u8|s16:channel_layouts=stereo
908	@end example
909
910	@section agate
911
912	A gate is mainly used to reduce lower parts of a signal. This kind of signal
913	processing reduces disturbing noise between useful signals.
914
915	Gating is done by detecting the volume below a chosen level @var{threshold}
916	and dividing it by the factor set with @var{ratio}. The bottom of the noise
917	floor is set via @var{range}. Because an exact manipulation of the signal
918	would cause distortion of the waveform the reduction can be levelled over
919	time. This is done by setting @var{attack} and @var{release}.
920
921	@var{attack} determines how long the signal has to fall below the threshold
922	before any reduction will occur and @var{release} sets the time the signal
923	has to rise above the threshold to reduce the reduction again.
924	Shorter signals than the chosen attack time will be left untouched.
925
926	@table @option
927	@item level_in
928	Set input level before filtering.
929	Default is 1. Allowed range is from 0.015625 to 64.
930
931	@item range
932	Set the level of gain reduction when the signal is below the threshold.
933	Default is 0.06125. Allowed range is from 0 to 1.
934
935	@item threshold
936	If a signal rises above this level the gain reduction is released.
937	Default is 0.125. Allowed range is from 0 to 1.
938
939	@item ratio
940	Set a ratio by which the signal is reduced.
941	Default is 2. Allowed range is from 1 to 9000.
942
943	@item attack
944	Amount of milliseconds the signal has to rise above the threshold before gain
945	reduction stops.
946	Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
947
948	@item release
949	Amount of milliseconds the signal has to fall below the threshold before the
950	reduction is increased again. Default is 250 milliseconds.
951	Allowed range is from 0.01 to 9000.
952
953	@item makeup
954	Set amount of amplification of signal after processing.
955	Default is 1. Allowed range is from 1 to 64.
956
957	@item knee
958	Curve the sharp knee around the threshold to enter gain reduction more softly.
959	Default is 2.828427125. Allowed range is from 1 to 8.
960
961	@item detection
962	Choose if exact signal should be taken for detection or an RMS like one.
963	Default is @code{rms}. Can be @code{peak} or @code{rms}.
964
965	@item link
966	Choose if the average level between all channels or the louder channel affects
967	the reduction.
968	Default is @code{average}. Can be @code{average} or @code{maximum}.
969	@end table
970
971	@section alimiter
972
973	The limiter prevents an input signal from rising over a desired threshold.
974	This limiter uses lookahead technology to prevent your signal from distorting.
975	It means that there is a small delay after the signal is processed. Keep in mind
976	that the delay it produces is the attack time you set.
977
978	The filter accepts the following options:
979
980	@table @option
981	@item level_in
982	Set input gain. Default is 1.
983
984	@item level_out
985	Set output gain. Default is 1.
986
987	@item limit
988	Don't let signals above this level pass the limiter. Default is 1.
989
990	@item attack
991	The limiter will reach its attenuation level in this amount of time in
992	milliseconds. Default is 5 milliseconds.
993
994	@item release
995	Come back from limiting to attenuation 1.0 in this amount of milliseconds.
996	Default is 50 milliseconds.
997
998	@item asc
999	When gain reduction is always needed ASC takes care of releasing to an
1000	average reduction level rather than reaching a reduction of 0 in the release
1001	time.
1002
1003	@item asc_level
1004	Select how much the release time is affected by ASC, 0 means nearly no changes
1005	in release time while 1 produces higher release times.
1006
1007	@item level
1008	Auto level output signal. Default is enabled.
1009	This normalizes audio back to 0dB if enabled.
1010	@end table
1011
1012	Depending on picked setting it is recommended to upsample input 2x or 4x times
1013	with @ref{aresample} before applying this filter.
1014
1015	@section allpass
1016
1017	Apply a two-pole all-pass filter with central frequency (in Hz)
1018	@var{frequency}, and filter-width @var{width}.
1019	An all-pass filter changes the audio's frequency to phase relationship
1020	without changing its frequency to amplitude relationship.
1021
1022	The filter accepts the following options:
1023
1024	@table @option
1025	@item frequency, f
1026	Set frequency in Hz.
1027
1028	@item width_type
1029	Set method to specify band-width of filter.
1030	@table @option
1031	@item h
1032	Hz
1033	@item q
1034	Q-Factor
1035	@item o
1036	octave
1037	@item s
1038	slope
1039	@end table
1040
1041	@item width, w
1042	Specify the band-width of a filter in width_type units.
1043	@end table
1044
1045	@section aloop
1046
1047	Loop audio samples.
1048
1049	The filter accepts the following options:
1050
1051	@table @option
1052	@item loop
1053	Set the number of loops.
1054
1055	@item size
1056	Set maximal number of samples.
1057
1058	@item start
1059	Set first sample of loop.
1060	@end table
1061
1062	@anchor{amerge}
1063	@section amerge
1064
1065	Merge two or more audio streams into a single multi-channel stream.
1066
1067	The filter accepts the following options:
1068
1069	@table @option
1070
1071	@item inputs
1072	Set the number of inputs. Default is 2.
1073
1074	@end table
1075
1076	If the channel layouts of the inputs are disjoint, and therefore compatible,
1077	the channel layout of the output will be set accordingly and the channels
1078	will be reordered as necessary. If the channel layouts of the inputs are not
1079	disjoint, the output will have all the channels of the first input then all
1080	the channels of the second input, in that order, and the channel layout of
1081	the output will be the default value corresponding to the total number of
1082	channels.
1083
1084	For example, if the first input is in 2.1 (FL+FR+LF) and the second input
1085	is FC+BL+BR, then the output will be in 5.1, with the channels in the
1086	following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
1087	first input, b1 is the first channel of the second input).
1088
1089	On the other hand, if both input are in stereo, the output channels will be
1090	in the default order: a1, a2, b1, b2, and the channel layout will be
1091	arbitrarily set to 4.0, which may or may not be the expected value.
1092
1093	All inputs must have the same sample rate, and format.
1094
1095	If inputs do not have the same duration, the output will stop with the
1096	shortest.
1097
1098	@subsection Examples
1099
1100	@itemize
1101	@item
1102	Merge two mono files into a stereo stream:
1103	@example
1104	amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
1105	@end example
1106
1107	@item
1108	Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
1109	@example
1110	ffmpeg -i input.mkv -filter_complex "[0:1][0:2][0:3][0:4][0:5][0:6] amerge=inputs=6" -c:a pcm_s16le output.mkv
1111	@end example
1112	@end itemize
1113
1114	@section amix
1115
1116	Mixes multiple audio inputs into a single output.
1117
1118	Note that this filter only supports float samples (the @var{amerge}
1119	and @var{pan} audio filters support many formats). If the @var{amix}
1120	input has integer samples then @ref{aresample} will be automatically
1121	inserted to perform the conversion to float samples.
1122
1123	For example
1124	@example
1125	ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
1126	@end example
1127	will mix 3 input audio streams to a single output with the same duration as the
1128	first input and a dropout transition time of 3 seconds.
1129
1130	It accepts the following parameters:
1131	@table @option
1132
1133	@item inputs
1134	The number of inputs. If unspecified, it defaults to 2.
1135
1136	@item duration
1137	How to determine the end-of-stream.
1138	@table @option
1139
1140	@item longest
1141	The duration of the longest input. (default)
1142
1143	@item shortest
1144	The duration of the shortest input.
1145
1146	@item first
1147	The duration of the first input.
1148
1149	@end table
1150
1151	@item dropout_transition
1152	The transition time, in seconds, for volume renormalization when an input
1153	stream ends. The default value is 2 seconds.
1154
1155	@end table
1156
1157	@section anequalizer
1158
1159	High-order parametric multiband equalizer for each channel.
1160
1161	It accepts the following parameters:
1162	@table @option
1163	@item params
1164
1165	This option string is in format:
1166	"c@var{chn} f=@var{cf} w=@var{w} g=@var{g} t=@var{f} | ..."
1167	Each equalizer band is separated by '|'.
1168
1169	@table @option
1170	@item chn
1171	Set channel number to which equalization will be applied.
1172	If input doesn't have that channel the entry is ignored.
1173
1174	@item f
1175	Set central frequency for band.
1176	If input doesn't have that frequency the entry is ignored.
1177
1178	@item w
1179	Set band width in hertz.
1180
1181	@item g
1182	Set band gain in dB.
1183
1184	@item t
1185	Set filter type for band, optional, can be:
1186
1187	@table @samp
1188	@item 0
1189	Butterworth, this is default.
1190
1191	@item 1
1192	Chebyshev type 1.
1193
1194	@item 2
1195	Chebyshev type 2.
1196	@end table
1197	@end table
1198
1199	@item curves
1200	With this option activated frequency response of anequalizer is displayed
1201	in video stream.
1202
1203	@item size
1204	Set video stream size. Only useful if curves option is activated.
1205
1206	@item mgain
1207	Set max gain that will be displayed. Only useful if curves option is activated.
1208	Setting this to a reasonable value makes it possible to display gain which is derived from
1209	neighbour bands which are too close to each other and thus produce higher gain
1210	when both are activated.
1211
1212	@item fscale
1213	Set frequency scale used to draw frequency response in video output.
1214	Can be linear or logarithmic. Default is logarithmic.
1215
1216	@item colors
1217	Set color for each channel curve which is going to be displayed in video stream.
1218	This is list of color names separated by space or by '|'.
1219	Unrecognised or missing colors will be replaced by white color.
1220	@end table
1221
1222	@subsection Examples
1223
1224	@itemize
1225	@item
1226	Lower gain by 10 of central frequency 200Hz and width 100 Hz
1227	for first 2 channels using Chebyshev type 1 filter:
1228	@example
1229	anequalizer=c0 f=200 w=100 g=-10 t=1|c1 f=200 w=100 g=-10 t=1
1230	@end example
1231	@end itemize
1232
1233	@subsection Commands
1234
1235	This filter supports the following commands:
1236	@table @option
1237	@item change
1238	Alter existing filter parameters.
1239	Syntax for the commands is : "@var{fN}|f=@var{freq}|w=@var{width}|g=@var{gain}"
1240
1241	@var{fN} is existing filter number, starting from 0, if no such filter is available
1242	error is returned.
1243	@var{freq} set new frequency parameter.
1244	@var{width} set new width parameter in herz.
1245	@var{gain} set new gain parameter in dB.
1246
1247	Full filter invocation with asendcmd may look like this:
1248	asendcmd=c='4.0 anequalizer change 0|f=200|w=50|g=1',anequalizer=...
1249	@end table
1250
1251	@section anull
1252
1253	Pass the audio source unchanged to the output.
1254
1255	@section apad
1256
1257	Pad the end of an audio stream with silence.
1258
1259	This can be used together with @command{ffmpeg} @option{-shortest} to
1260	extend audio streams to the same length as the video stream.
1261
1262	A description of the accepted options follows.
1263
1264	@table @option
1265	@item packet_size
1266	Set silence packet size. Default value is 4096.
1267
1268	@item pad_len
1269	Set the number of samples of silence to add to the end. After the
1270	value is reached, the stream is terminated. This option is mutually
1271	exclusive with @option{whole_len}.
1272
1273	@item whole_len
1274	Set the minimum total number of samples in the output audio stream. If
1275	the value is longer than the input audio length, silence is added to
1276	the end, until the value is reached. This option is mutually exclusive
1277	with @option{pad_len}.
1278	@end table
1279
1280	If neither the @option{pad_len} nor the @option{whole_len} option is
1281	set, the filter will add silence to the end of the input stream
1282	indefinitely.
1283
1284	@subsection Examples
1285
1286	@itemize
1287	@item
1288	Add 1024 samples of silence to the end of the input:
1289	@example
1290	apad=pad_len=1024
1291	@end example
1292
1293	@item
1294	Make sure the audio output will contain at least 10000 samples, pad
1295	the input with silence if required:
1296	@example
1297	apad=whole_len=10000
1298	@end example
1299
1300	@item
1301	Use @command{ffmpeg} to pad the audio input with silence, so that the
1302	video stream will always result the shortest and will be converted
1303	until the end in the output file when using the @option{shortest}
1304	option:
1305	@example
1306	ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
1307	@end example
1308	@end itemize
1309
1310	@section aphaser
1311	Add a phasing effect to the input audio.
1312
1313	A phaser filter creates series of peaks and troughs in the frequency spectrum.
1314	The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
1315
1316	A description of the accepted parameters follows.
1317
1318	@table @option
1319	@item in_gain
1320	Set input gain. Default is 0.4.
1321
1322	@item out_gain
1323	Set output gain. Default is 0.74
1324
1325	@item delay
1326	Set delay in milliseconds. Default is 3.0.
1327
1328	@item decay
1329	Set decay. Default is 0.4.
1330
1331	@item speed
1332	Set modulation speed in Hz. Default is 0.5.
1333
1334	@item type
1335	Set modulation type. Default is triangular.
1336
1337	It accepts the following values:
1338	@table @samp
1339	@item triangular, t
1340	@item sinusoidal, s
1341	@end table
1342	@end table
1343
1344	@section apulsator
1345
1346	Audio pulsator is something between an autopanner and a tremolo.
1347	But it can produce funny stereo effects as well. Pulsator changes the volume
1348	of the left and right channel based on a LFO (low frequency oscillator) with
1349	different waveforms and shifted phases.
1350	This filter have the ability to define an offset between left and right
1351	channel. An offset of 0 means that both LFO shapes match each other.
1352	The left and right channel are altered equally - a conventional tremolo.
1353	An offset of 50% means that the shape of the right channel is exactly shifted
1354	in phase (or moved backwards about half of the frequency) - pulsator acts as
1355	an autopanner. At 1 both curves match again. Every setting in between moves the
1356	phase shift gapless between all stages and produces some "bypassing" sounds with
1357	sine and triangle waveforms. The more you set the offset near 1 (starting from
1358	the 0.5) the faster the signal passes from the left to the right speaker.
1359
1360	The filter accepts the following options:
1361
1362	@table @option
1363	@item level_in
1364	Set input gain. By default it is 1. Range is [0.015625 - 64].
1365
1366	@item level_out
1367	Set output gain. By default it is 1. Range is [0.015625 - 64].
1368
1369	@item mode
1370	Set waveform shape the LFO will use. Can be one of: sine, triangle, square,
1371	sawup or sawdown. Default is sine.
1372
1373	@item amount
1374	Set modulation. Define how much of original signal is affected by the LFO.
1375
1376	@item offset_l
1377	Set left channel offset. Default is 0. Allowed range is [0 - 1].
1378
1379	@item offset_r
1380	Set right channel offset. Default is 0.5. Allowed range is [0 - 1].
1381
1382	@item width
1383	Set pulse width. Default is 1. Allowed range is [0 - 2].
1384
1385	@item timing
1386	Set possible timing mode. Can be one of: bpm, ms or hz. Default is hz.
1387
1388	@item bpm
1389	Set bpm. Default is 120. Allowed range is [30 - 300]. Only used if timing
1390	is set to bpm.
1391
1392	@item ms
1393	Set ms. Default is 500. Allowed range is [10 - 2000]. Only used if timing
1394	is set to ms.
1395
1396	@item hz
1397	Set frequency in Hz. Default is 2. Allowed range is [0.01 - 100]. Only used
1398	if timing is set to hz.
1399	@end table
1400
1401	@anchor{aresample}
1402	@section aresample
1403
1404	Resample the input audio to the specified parameters, using the
1405	libswresample library. If none are specified then the filter will
1406	automatically convert between its input and output.
1407
1408	This filter is also able to stretch/squeeze the audio data to make it match
1409	the timestamps or to inject silence / cut out audio to make it match the
1410	timestamps, do a combination of both or do neither.
1411
1412	The filter accepts the syntax
1413	[@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
1414	expresses a sample rate and @var{resampler_options} is a list of
1415	@var{key}=@var{value} pairs, separated by ":". See the
1416	@ref{Resampler Options,,the "Resampler Options" section in the
1417	ffmpeg-resampler(1) manual,ffmpeg-resampler}
1418	for the complete list of supported options.
1419
1420	@subsection Examples
1421
1422	@itemize
1423	@item
1424	Resample the input audio to 44100Hz:
1425	@example
1426	aresample=44100
1427	@end example
1428
1429	@item
1430	Stretch/squeeze samples to the given timestamps, with a maximum of 1000
1431	samples per second compensation:
1432	@example
1433	aresample=async=1000
1434	@end example
1435	@end itemize
1436
1437	@section areverse
1438
1439	Reverse an audio clip.
1440
1441	Warning: This filter requires memory to buffer the entire clip, so trimming
1442	is suggested.
1443
1444	@subsection Examples
1445
1446	@itemize
1447	@item
1448	Take the first 5 seconds of a clip, and reverse it.
1449	@example
1450	atrim=end=5,areverse
1451	@end example
1452	@end itemize
1453
1454	@section asetnsamples
1455
1456	Set the number of samples per each output audio frame.
1457
1458	The last output packet may contain a different number of samples, as
1459	the filter will flush all the remaining samples when the input audio
1460	signals its end.
1461
1462	The filter accepts the following options:
1463
1464	@table @option
1465
1466	@item nb_out_samples, n
1467	Set the number of frames per each output audio frame. The number is
1468	intended as the number of samples @emph{per each channel}.
1469	Default value is 1024.
1470
1471	@item pad, p
1472	If set to 1, the filter will pad the last audio frame with zeroes, so
1473	that the last frame will contain the same number of samples as the
1474	previous ones. Default value is 1.
1475	@end table
1476
1477	For example, to set the number of per-frame samples to 1234 and
1478	disable padding for the last frame, use:
1479	@example
1480	asetnsamples=n=1234:p=0
1481	@end example
1482
1483	@section asetrate
1484
1485	Set the sample rate without altering the PCM data.
1486	This will result in a change of speed and pitch.
1487
1488	The filter accepts the following options:
1489
1490	@table @option
1491	@item sample_rate, r
1492	Set the output sample rate. Default is 44100 Hz.
1493	@end table
1494
1495	@section ashowinfo
1496
1497	Show a line containing various information for each input audio frame.
1498	The input audio is not modified.
1499
1500	The shown line contains a sequence of key/value pairs of the form
1501	@var{key}:@var{value}.
1502
1503	The following values are shown in the output:
1504
1505	@table @option
1506	@item n
1507	The (sequential) number of the input frame, starting from 0.
1508
1509	@item pts
1510	The presentation timestamp of the input frame, in time base units; the time base
1511	depends on the filter input pad, and is usually 1/@var{sample_rate}.
1512
1513	@item pts_time
1514	The presentation timestamp of the input frame in seconds.
1515
1516	@item pos
1517	position of the frame in the input stream, -1 if this information in
1518	unavailable and/or meaningless (for example in case of synthetic audio)
1519
1520	@item fmt
1521	The sample format.
1522
1523	@item chlayout
1524	The channel layout.
1525
1526	@item rate
1527	The sample rate for the audio frame.
1528
1529	@item nb_samples
1530	The number of samples (per channel) in the frame.
1531
1532	@item checksum
1533	The Adler-32 checksum (printed in hexadecimal) of the audio data. For planar
1534	audio, the data is treated as if all the planes were concatenated.
1535
1536	@item plane_checksums
1537	A list of Adler-32 checksums for each data plane.
1538	@end table
1539
1540	@anchor{astats}
1541	@section astats
1542
1543	Display time domain statistical information about the audio channels.
1544	Statistics are calculated and displayed for each audio channel and,
1545	where applicable, an overall figure is also given.
1546
1547	It accepts the following option:
1548	@table @option
1549	@item length
1550	Short window length in seconds, used for peak and trough RMS measurement.
1551	Default is @code{0.05} (50 milliseconds). Allowed range is @code{[0.1 - 10]}.
1552
1553	@item metadata
1554
1555	Set metadata injection. All the metadata keys are prefixed with @code{lavfi.astats.X},
1556	where @code{X} is channel number starting from 1 or string @code{Overall}. Default is
1557	disabled.
1558
1559	Available keys for each channel are:
1560	DC_offset
1561	Min_level
1562	Max_level
1563	Min_difference
1564	Max_difference
1565	Mean_difference
1566	Peak_level
1567	RMS_peak
1568	RMS_trough
1569	Crest_factor
1570	Flat_factor
1571	Peak_count
1572	Bit_depth
1573
1574	and for Overall:
1575	DC_offset
1576	Min_level
1577	Max_level
1578	Min_difference
1579	Max_difference
1580	Mean_difference
1581	Peak_level
1582	RMS_level
1583	RMS_peak
1584	RMS_trough
1585	Flat_factor
1586	Peak_count
1587	Bit_depth
1588	Number_of_samples
1589
1590	For example full key look like this @code{lavfi.astats.1.DC_offset} or
1591	this @code{lavfi.astats.Overall.Peak_count}.
1592
1593	For description what each key means read below.
1594
1595	@item reset
1596	Set number of frame after which stats are going to be recalculated.
1597	Default is disabled.
1598	@end table
1599
1600	A description of each shown parameter follows:
1601
1602	@table @option
1603	@item DC offset
1604	Mean amplitude displacement from zero.
1605
1606	@item Min level
1607	Minimal sample level.
1608
1609	@item Max level
1610	Maximal sample level.
1611
1612	@item Min difference
1613	Minimal difference between two consecutive samples.
1614
1615	@item Max difference
1616	Maximal difference between two consecutive samples.
1617
1618	@item Mean difference
1619	Mean difference between two consecutive samples.
1620	The average of each difference between two consecutive samples.
1621
1622	@item Peak level dB
1623	@item RMS level dB
1624	Standard peak and RMS level measured in dBFS.
1625
1626	@item RMS peak dB
1627	@item RMS trough dB
1628	Peak and trough values for RMS level measured over a short window.
1629
1630	@item Crest factor
1631	Standard ratio of peak to RMS level (note: not in dB).
1632
1633	@item Flat factor
1634	Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
1635	(i.e. either @var{Min level} or @var{Max level}).
1636
1637	@item Peak count
1638	Number of occasions (not the number of samples) that the signal attained either
1639	@var{Min level} or @var{Max level}.
1640
1641	@item Bit depth
1642	Overall bit depth of audio. Number of bits used for each sample.
1643	@end table
1644
1645	@section atempo
1646
1647	Adjust audio tempo.
1648
1649	The filter accepts exactly one parameter, the audio tempo. If not
1650	specified then the filter will assume nominal 1.0 tempo. Tempo must
1651	be in the [0.5, 2.0] range.
1652
1653	@subsection Examples
1654
1655	@itemize
1656	@item
1657	Slow down audio to 80% tempo:
1658	@example
1659	atempo=0.8
1660	@end example
1661
1662	@item
1663	To speed up audio to 125% tempo:
1664	@example
1665	atempo=1.25
1666	@end example
1667	@end itemize
1668
1669	@section atrim
1670
1671	Trim the input so that the output contains one continuous subpart of the input.
1672
1673	It accepts the following parameters:
1674	@table @option
1675	@item start
1676	Timestamp (in seconds) of the start of the section to keep. I.e. the audio
1677	sample with the timestamp @var{start} will be the first sample in the output.
1678
1679	@item end
1680	Specify time of the first audio sample that will be dropped, i.e. the
1681	audio sample immediately preceding the one with the timestamp @var{end} will be
1682	the last sample in the output.
1683
1684	@item start_pts
1685	Same as @var{start}, except this option sets the start timestamp in samples
1686	instead of seconds.
1687
1688	@item end_pts
1689	Same as @var{end}, except this option sets the end timestamp in samples instead
1690	of seconds.
1691
1692	@item duration
1693	The maximum duration of the output in seconds.
1694
1695	@item start_sample
1696	The number of the first sample that should be output.
1697
1698	@item end_sample
1699	The number of the first sample that should be dropped.
1700	@end table
1701
1702	@option{start}, @option{end}, and @option{duration} are expressed as time
1703	duration specifications; see
1704	@ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
1705
1706	Note that the first two sets of the start/end options and the @option{duration}
1707	option look at the frame timestamp, while the _sample options simply count the
1708	samples that pass through the filter. So start/end_pts and start/end_sample will
1709	give different results when the timestamps are wrong, inexact or do not start at
1710	zero. Also note that this filter does not modify the timestamps. If you wish
1711	to have the output timestamps start at zero, insert the asetpts filter after the
1712	atrim filter.
1713
1714	If multiple start or end options are set, this filter tries to be greedy and
1715	keep all samples that match at least one of the specified constraints. To keep
1716	only the part that matches all the constraints at once, chain multiple atrim
1717	filters.
1718
1719	The defaults are such that all the input is kept. So it is possible to set e.g.
1720	just the end values to keep everything before the specified time.
1721
1722	Examples:
1723	@itemize
1724	@item
1725	Drop everything except the second minute of input:
1726	@example
1727	ffmpeg -i INPUT -af atrim=60:120
1728	@end example
1729
1730	@item
1731	Keep only the first 1000 samples:
1732	@example
1733	ffmpeg -i INPUT -af atrim=end_sample=1000
1734	@end example
1735
1736	@end itemize
1737
1738	@section bandpass
1739
1740	Apply a two-pole Butterworth band-pass filter with central
1741	frequency @var{frequency}, and (3dB-point) band-width width.
1742	The @var{csg} option selects a constant skirt gain (peak gain = Q)
1743	instead of the default: constant 0dB peak gain.
1744	The filter roll off at 6dB per octave (20dB per decade).
1745
1746	The filter accepts the following options:
1747
1748	@table @option
1749	@item frequency, f
1750	Set the filter's central frequency. Default is @code{3000}.
1751
1752	@item csg
1753	Constant skirt gain if set to 1. Defaults to 0.
1754
1755	@item width_type
1756	Set method to specify band-width of filter.
1757	@table @option
1758	@item h
1759	Hz
1760	@item q
1761	Q-Factor
1762	@item o
1763	octave
1764	@item s
1765	slope
1766	@end table
1767
1768	@item width, w
1769	Specify the band-width of a filter in width_type units.
1770	@end table
1771
1772	@section bandreject
1773
1774	Apply a two-pole Butterworth band-reject filter with central
1775	frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1776	The filter roll off at 6dB per octave (20dB per decade).
1777
1778	The filter accepts the following options:
1779
1780	@table @option
1781	@item frequency, f
1782	Set the filter's central frequency. Default is @code{3000}.
1783
1784	@item width_type
1785	Set method to specify band-width of filter.
1786	@table @option
1787	@item h
1788	Hz
1789	@item q
1790	Q-Factor
1791	@item o
1792	octave
1793	@item s
1794	slope
1795	@end table
1796
1797	@item width, w
1798	Specify the band-width of a filter in width_type units.
1799	@end table
1800
1801	@section bass
1802
1803	Boost or cut the bass (lower) frequencies of the audio using a two-pole
1804	shelving filter with a response similar to that of a standard
1805	hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1806
1807	The filter accepts the following options:
1808
1809	@table @option
1810	@item gain, g
1811	Give the gain at 0 Hz. Its useful range is about -20
1812	(for a large cut) to +20 (for a large boost).
1813	Beware of clipping when using a positive gain.
1814
1815	@item frequency, f
1816	Set the filter's central frequency and so can be used
1817	to extend or reduce the frequency range to be boosted or cut.
1818	The default value is @code{100} Hz.
1819
1820	@item width_type
1821	Set method to specify band-width of filter.
1822	@table @option
1823	@item h
1824	Hz
1825	@item q
1826	Q-Factor
1827	@item o
1828	octave
1829	@item s
1830	slope
1831	@end table
1832
1833	@item width, w
1834	Determine how steep is the filter's shelf transition.
1835	@end table
1836
1837	@section biquad
1838
1839	Apply a biquad IIR filter with the given coefficients.
1840	Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1841	are the numerator and denominator coefficients respectively.
1842
1843	@section bs2b
1844	Bauer stereo to binaural transformation, which improves headphone listening of
1845	stereo audio records.
1846
1847	It accepts the following parameters:
1848	@table @option
1849
1850	@item profile
1851	Pre-defined crossfeed level.
1852	@table @option
1853
1854	@item default
1855	Default level (fcut=700, feed=50).
1856
1857	@item cmoy
1858	Chu Moy circuit (fcut=700, feed=60).
1859
1860	@item jmeier
1861	Jan Meier circuit (fcut=650, feed=95).
1862
1863	@end table
1864
1865	@item fcut
1866	Cut frequency (in Hz).
1867
1868	@item feed
1869	Feed level (in Hz).
1870
1871	@end table
1872
1873	@section channelmap
1874
1875	Remap input channels to new locations.
1876
1877	It accepts the following parameters:
1878	@table @option
1879	@item map
1880	Map channels from input to output. The argument is a '|'-separated list of
1881	mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1882	@var{in_channel} form. @var{in_channel} can be either the name of the input
1883	channel (e.g. FL for front left) or its index in the input channel layout.
1884	@var{out_channel} is the name of the output channel or its index in the output
1885	channel layout. If @var{out_channel} is not given then it is implicitly an
1886	index, starting with zero and increasing by one for each mapping.
1887
1888	@item channel_layout
1889	The channel layout of the output stream.
1890	@end table
1891
1892	If no mapping is present, the filter will implicitly map input channels to
1893	output channels, preserving indices.
1894
1895	For example, assuming a 5.1+downmix input MOV file,
1896	@example
1897	ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1898	@end example
1899	will create an output WAV file tagged as stereo from the downmix channels of
1900	the input.
1901
1902	To fix a 5.1 WAV improperly encoded in AAC's native channel order
1903	@example
1904	ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:5.1' out.wav
1905	@end example
1906
1907	@section channelsplit
1908
1909	Split each channel from an input audio stream into a separate output stream.
1910
1911	It accepts the following parameters:
1912	@table @option
1913	@item channel_layout
1914	The channel layout of the input stream. The default is "stereo".
1915	@end table
1916
1917	For example, assuming a stereo input MP3 file,
1918	@example
1919	ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1920	@end example
1921	will create an output Matroska file with two audio streams, one containing only
1922	the left channel and the other the right channel.
1923
1924	Split a 5.1 WAV file into per-channel files:
1925	@example
1926	ffmpeg -i in.wav -filter_complex
1927	'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1928	-map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1929	front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1930	side_right.wav
1931	@end example
1932
1933	@section chorus
1934	Add a chorus effect to the audio.
1935
1936	Can make a single vocal sound like a chorus, but can also be applied to instrumentation.
1937
1938	Chorus resembles an echo effect with a short delay, but whereas with echo the delay is
1939	constant, with chorus, it is varied using using sinusoidal or triangular modulation.
1940	The modulation depth defines the range the modulated delay is played before or after
1941	the delay. Hence the delayed sound will sound slower or faster, that is the delayed
1942	sound tuned around the original one, like in a chorus where some vocals are slightly
1943	off key.
1944
1945	It accepts the following parameters:
1946	@table @option
1947	@item in_gain
1948	Set input gain. Default is 0.4.
1949
1950	@item out_gain
1951	Set output gain. Default is 0.4.
1952
1953	@item delays
1954	Set delays. A typical delay is around 40ms to 60ms.
1955
1956	@item decays
1957	Set decays.
1958
1959	@item speeds
1960	Set speeds.
1961
1962	@item depths
1963	Set depths.
1964	@end table
1965
1966	@subsection Examples
1967
1968	@itemize
1969	@item
1970	A single delay:
1971	@example
1972	chorus=0.7:0.9:55:0.4:0.25:2
1973	@end example
1974
1975	@item
1976	Two delays:
1977	@example
1978	chorus=0.6:0.9:50|60:0.4|0.32:0.25|0.4:2|1.3
1979	@end example
1980
1981	@item
1982	Fuller sounding chorus with three delays:
1983	@example
1984	chorus=0.5:0.9:50|60|40:0.4|0.32|0.3:0.25|0.4|0.3:2|2.3|1.3
1985	@end example
1986	@end itemize
1987
1988	@section compand
1989	Compress or expand the audio's dynamic range.
1990
1991	It accepts the following parameters:
1992
1993	@table @option
1994
1995	@item attacks
1996	@item decays
1997	A list of times in seconds for each channel over which the instantaneous level
1998	of the input signal is averaged to determine its volume. @var{attacks} refers to
1999	increase of volume and @var{decays} refers to decrease of volume. For most
2000	situations, the attack time (response to the audio getting louder) should be
2001	shorter than the decay time, because the human ear is more sensitive to sudden
2002	loud audio than sudden soft audio. A typical value for attack is 0.3 seconds and
2003	a typical value for decay is 0.8 seconds.
2004	If specified number of attacks & decays is lower than number of channels, the last
2005	set attack/decay will be used for all remaining channels.
2006
2007	@item points
2008	A list of points for the transfer function, specified in dB relative to the
2009	maximum possible signal amplitude. Each key points list must be defined using
2010	the following syntax: @code{x0/y0|x1/y1|x2/y2|....} or
2011	@code{x0/y0 x1/y1 x2/y2 ....}
2012
2013	The input values must be in strictly increasing order but the transfer function
2014	does not have to be monotonically rising. The point @code{0/0} is assumed but
2015	may be overridden (by @code{0/out-dBn}). Typical values for the transfer
2016	function are @code{-70/-70|-60/-20}.
2017
2018	@item soft-knee
2019	Set the curve radius in dB for all joints. It defaults to 0.01.
2020
2021	@item gain
2022	Set the additional gain in dB to be applied at all points on the transfer
2023	function. This allows for easy adjustment of the overall gain.
2024	It defaults to 0.
2025
2026	@item volume
2027	Set an initial volume, in dB, to be assumed for each channel when filtering
2028	starts. This permits the user to supply a nominal level initially, so that, for
2029	example, a very large gain is not applied to initial signal levels before the
2030	companding has begun to operate. A typical value for audio which is initially
2031	quiet is -90 dB. It defaults to 0.
2032
2033	@item delay
2034	Set a delay, in seconds. The input audio is analyzed immediately, but audio is
2035	delayed before being fed to the volume adjuster. Specifying a delay
2036	approximately equal to the attack/decay times allows the filter to effectively
2037	operate in predictive rather than reactive mode. It defaults to 0.
2038
2039	@end table
2040
2041	@subsection Examples
2042
2043	@itemize
2044	@item
2045	Make music with both quiet and loud passages suitable for listening to in a
2046	noisy environment:
2047	@example
2048	compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
2049	@end example
2050
2051	Another example for audio with whisper and explosion parts:
2052	@example
2053	compand=0|0:1|1:-90/-900|-70/-70|-30/-9|0/-3:6:0:0:0
2054	@end example
2055
2056	@item
2057	A noise gate for when the noise is at a lower level than the signal:
2058	@example
2059	compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
2060	@end example
2061
2062	@item
2063	Here is another noise gate, this time for when the noise is at a higher level
2064	than the signal (making it, in some ways, similar to squelch):
2065	@example
2066	compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
2067	@end example
2068
2069	@item
2070	2:1 compression starting at -6dB:
2071	@example
2072	compand=points=-80/-80|-6/-6|0/-3.8|20/3.5
2073	@end example
2074
2075	@item
2076	2:1 compression starting at -9dB:
2077	@example
2078	compand=points=-80/-80|-9/-9|0/-5.3|20/2.9
2079	@end example
2080
2081	@item
2082	2:1 compression starting at -12dB:
2083	@example
2084	compand=points=-80/-80|-12/-12|0/-6.8|20/1.9
2085	@end example
2086
2087	@item
2088	2:1 compression starting at -18dB:
2089	@example
2090	compand=points=-80/-80|-18/-18|0/-9.8|20/0.7
2091	@end example
2092
2093	@item
2094	3:1 compression starting at -15dB:
2095	@example
2096	compand=points=-80/-80|-15/-15|0/-10.8|20/-5.2
2097	@end example
2098
2099	@item
2100	Compressor/Gate:
2101	@example
2102	compand=points=-80/-105|-62/-80|-15.4/-15.4|0/-12|20/-7.6
2103	@end example
2104
2105	@item
2106	Expander:
2107	@example
2108	compand=attacks=0:points=-80/-169|-54/-80|-49.5/-64.6|-41.1/-41.1|-25.8/-15|-10.8/-4.5|0/0|20/8.3
2109	@end example
2110
2111	@item
2112	Hard limiter at -6dB:
2113	@example
2114	compand=attacks=0:points=-80/-80|-6/-6|20/-6
2115	@end example
2116
2117	@item
2118	Hard limiter at -12dB:
2119	@example
2120	compand=attacks=0:points=-80/-80|-12/-12|20/-12
2121	@end example
2122
2123	@item
2124	Hard noise gate at -35 dB:
2125	@example
2126	compand=attacks=0:points=-80/-115|-35.1/-80|-35/-35|20/20
2127	@end example
2128
2129	@item
2130	Soft limiter:
2131	@example
2132	compand=attacks=0:points=-80/-80|-12.4/-12.4|-6/-8|0/-6.8|20/-2.8
2133	@end example
2134	@end itemize
2135
2136	@section compensationdelay
2137
2138	Compensation Delay Line is a metric based delay to compensate differing
2139	positions of microphones or speakers.
2140
2141	For example, you have recorded guitar with two microphones placed in
2142	different location. Because the front of sound wave has fixed speed in
2143	normal conditions, the phasing of microphones can vary and depends on
2144	their location and interposition. The best sound mix can be achieved when
2145	these microphones are in phase (synchronized). Note that distance of
2146	~30 cm between microphones makes one microphone to capture signal in
2147	antiphase to another microphone. That makes the final mix sounding moody.
2148	This filter helps to solve phasing problems by adding different delays
2149	to each microphone track and make them synchronized.
2150
2151	The best result can be reached when you take one track as base and
2152	synchronize other tracks one by one with it.
2153	Remember that synchronization/delay tolerance depends on sample rate, too.
2154	Higher sample rates will give more tolerance.
2155
2156	It accepts the following parameters:
2157
2158	@table @option
2159	@item mm
2160	Set millimeters distance. This is compensation distance for fine tuning.
2161	Default is 0.
2162
2163	@item cm
2164	Set cm distance. This is compensation distance for tightening distance setup.
2165	Default is 0.
2166
2167	@item m
2168	Set meters distance. This is compensation distance for hard distance setup.
2169	Default is 0.
2170
2171	@item dry
2172	Set dry amount. Amount of unprocessed (dry) signal.
2173	Default is 0.
2174
2175	@item wet
2176	Set wet amount. Amount of processed (wet) signal.
2177	Default is 1.
2178
2179	@item temp
2180	Set temperature degree in Celsius. This is the temperature of the environment.
2181	Default is 20.
2182	@end table
2183
2184	@section crystalizer
2185	Simple algorithm to expand audio dynamic range.
2186
2187	The filter accepts the following options:
2188
2189	@table @option
2190	@item i
2191	Sets the intensity of effect (default: 2.0). Must be in range between 0.0
2192	(unchanged sound) to 10.0 (maximum effect).
2193
2194	@item c
2195	Enable clipping. By default is enabled.
2196	@end table
2197
2198	@section dcshift
2199	Apply a DC shift to the audio.
2200
2201	This can be useful to remove a DC offset (caused perhaps by a hardware problem
2202	in the recording chain) from the audio. The effect of a DC offset is reduced
2203	headroom and hence volume. The @ref{astats} filter can be used to determine if
2204	a signal has a DC offset.
2205
2206	@table @option
2207	@item shift
2208	Set the DC shift, allowed range is [-1, 1]. It indicates the amount to shift
2209	the audio.
2210
2211	@item limitergain
2212	Optional. It should have a value much less than 1 (e.g. 0.05 or 0.02) and is
2213	used to prevent clipping.
2214	@end table
2215
2216	@section dynaudnorm
2217	Dynamic Audio Normalizer.
2218
2219	This filter applies a certain amount of gain to the input audio in order
2220	to bring its peak magnitude to a target level (e.g. 0 dBFS). However, in
2221	contrast to more "simple" normalization algorithms, the Dynamic Audio
2222	Normalizer *dynamically* re-adjusts the gain factor to the input audio.
2223	This allows for applying extra gain to the "quiet" sections of the audio
2224	while avoiding distortions or clipping the "loud" sections. In other words:
2225	The Dynamic Audio Normalizer will "even out" the volume of quiet and loud
2226	sections, in the sense that the volume of each section is brought to the
2227	same target level. Note, however, that the Dynamic Audio Normalizer achieves
2228	this goal *without* applying "dynamic range compressing". It will retain 100%
2229	of the dynamic range *within* each section of the audio file.
2230
2231	@table @option
2232	@item f
2233	Set the frame length in milliseconds. In range from 10 to 8000 milliseconds.
2234	Default is 500 milliseconds.
2235	The Dynamic Audio Normalizer processes the input audio in small chunks,
2236	referred to as frames. This is required, because a peak magnitude has no
2237	meaning for just a single sample value. Instead, we need to determine the
2238	peak magnitude for a contiguous sequence of sample values. While a "standard"
2239	normalizer would simply use the peak magnitude of the complete file, the
2240	Dynamic Audio Normalizer determines the peak magnitude individually for each
2241	frame. The length of a frame is specified in milliseconds. By default, the
2242	Dynamic Audio Normalizer uses a frame length of 500 milliseconds, which has
2243	been found to give good results with most files.
2244	Note that the exact frame length, in number of samples, will be determined
2245	automatically, based on the sampling rate of the individual input audio file.
2246
2247	@item g
2248	Set the Gaussian filter window size. In range from 3 to 301, must be odd
2249	number. Default is 31.
2250	Probably the most important parameter of the Dynamic Audio Normalizer is the
2251	@code{window size} of the Gaussian smoothing filter. The filter's window size
2252	is specified in frames, centered around the current frame. For the sake of
2253	simplicity, this must be an odd number. Consequently, the default value of 31
2254	takes into account the current frame, as well as the 15 preceding frames and
2255	the 15 subsequent frames. Using a larger window results in a stronger
2256	smoothing effect and thus in less gain variation, i.e. slower gain
2257	adaptation. Conversely, using a smaller window results in a weaker smoothing
2258	effect and thus in more gain variation, i.e. faster gain adaptation.
2259	In other words, the more you increase this value, the more the Dynamic Audio
2260	Normalizer will behave like a "traditional" normalization filter. On the
2261	contrary, the more you decrease this value, the more the Dynamic Audio
2262	Normalizer will behave like a dynamic range compressor.
2263
2264	@item p
2265	Set the target peak value. This specifies the highest permissible magnitude
2266	level for the normalized audio input. This filter will try to approach the
2267	target peak magnitude as closely as possible, but at the same time it also
2268	makes sure that the normalized signal will never exceed the peak magnitude.
2269	A frame's maximum local gain factor is imposed directly by the target peak
2270	magnitude. The default value is 0.95 and thus leaves a headroom of 5%*.
2271	It is not recommended to go above this value.
2272
2273	@item m
2274	Set the maximum gain factor. In range from 1.0 to 100.0. Default is 10.0.
2275	The Dynamic Audio Normalizer determines the maximum possible (local) gain
2276	factor for each input frame, i.e. the maximum gain factor that does not
2277	result in clipping or distortion. The maximum gain factor is determined by
2278	the frame's highest magnitude sample. However, the Dynamic Audio Normalizer
2279	additionally bounds the frame's maximum gain factor by a predetermined
2280	(global) maximum gain factor. This is done in order to avoid excessive gain
2281	factors in "silent" or almost silent frames. By default, the maximum gain
2282	factor is 10.0, For most inputs the default value should be sufficient and
2283	it usually is not recommended to increase this value. Though, for input
2284	with an extremely low overall volume level, it may be necessary to allow even
2285	higher gain factors. Note, however, that the Dynamic Audio Normalizer does
2286	not simply apply a "hard" threshold (i.e. cut off values above the threshold).
2287	Instead, a "sigmoid" threshold function will be applied. This way, the
2288	gain factors will smoothly approach the threshold value, but never exceed that
2289	value.
2290
2291	@item r
2292	Set the target RMS. In range from 0.0 to 1.0. Default is 0.0 - disabled.
2293	By default, the Dynamic Audio Normalizer performs "peak" normalization.
2294	This means that the maximum local gain factor for each frame is defined
2295	(only) by the frame's highest magnitude sample. This way, the samples can
2296	be amplified as much as possible without exceeding the maximum signal
2297	level, i.e. without clipping. Optionally, however, the Dynamic Audio
2298	Normalizer can also take into account the frame's root mean square,
2299	abbreviated RMS. In electrical engineering, the RMS is commonly used to
2300	determine the power of a time-varying signal. It is therefore considered
2301	that the RMS is a better approximation of the "perceived loudness" than
2302	just looking at the signal's peak magnitude. Consequently, by adjusting all
2303	frames to a constant RMS value, a uniform "perceived loudness" can be
2304	established. If a target RMS value has been specified, a frame's local gain
2305	factor is defined as the factor that would result in exactly that RMS value.
2306	Note, however, that the maximum local gain factor is still restricted by the
2307	frame's highest magnitude sample, in order to prevent clipping.
2308
2309	@item n
2310	Enable channels coupling. By default is enabled.
2311	By default, the Dynamic Audio Normalizer will amplify all channels by the same
2312	amount. This means the same gain factor will be applied to all channels, i.e.
2313	the maximum possible gain factor is determined by the "loudest" channel.
2314	However, in some recordings, it may happen that the volume of the different
2315	channels is uneven, e.g. one channel may be "quieter" than the other one(s).
2316	In this case, this option can be used to disable the channel coupling. This way,
2317	the gain factor will be determined independently for each channel, depending
2318	only on the individual channel's highest magnitude sample. This allows for
2319	harmonizing the volume of the different channels.
2320
2321	@item c
2322	Enable DC bias correction. By default is disabled.
2323	An audio signal (in the time domain) is a sequence of sample values.
2324	In the Dynamic Audio Normalizer these sample values are represented in the
2325	-1.0 to 1.0 range, regardless of the original input format. Normally, the
2326	audio signal, or "waveform", should be centered around the zero point.
2327	That means if we calculate the mean value of all samples in a file, or in a
2328	single frame, then the result should be 0.0 or at least very close to that
2329	value. If, however, there is a significant deviation of the mean value from
2330	0.0, in either positive or negative direction, this is referred to as a
2331	DC bias or DC offset. Since a DC bias is clearly undesirable, the Dynamic
2332	Audio Normalizer provides optional DC bias correction.
2333	With DC bias correction enabled, the Dynamic Audio Normalizer will determine
2334	the mean value, or "DC correction" offset, of each input frame and subtract
2335	that value from all of the frame's sample values which ensures those samples
2336	are centered around 0.0 again. Also, in order to avoid "gaps" at the frame
2337	boundaries, the DC correction offset values will be interpolated smoothly
2338	between neighbouring frames.
2339
2340	@item b
2341	Enable alternative boundary mode. By default is disabled.
2342	The Dynamic Audio Normalizer takes into account a certain neighbourhood
2343	around each frame. This includes the preceding frames as well as the
2344	subsequent frames. However, for the "boundary" frames, located at the very
2345	beginning and at the very end of the audio file, not all neighbouring
2346	frames are available. In particular, for the first few frames in the audio
2347	file, the preceding frames are not known. And, similarly, for the last few
2348	frames in the audio file, the subsequent frames are not known. Thus, the
2349	question arises which gain factors should be assumed for the missing frames
2350	in the "boundary" region. The Dynamic Audio Normalizer implements two modes
2351	to deal with this situation. The default boundary mode assumes a gain factor
2352	of exactly 1.0 for the missing frames, resulting in a smooth "fade in" and
2353	"fade out" at the beginning and at the end of the input, respectively.
2354
2355	@item s
2356	Set the compress factor. In range from 0.0 to 30.0. Default is 0.0.
2357	By default, the Dynamic Audio Normalizer does not apply "traditional"
2358	compression. This means that signal peaks will not be pruned and thus the
2359	full dynamic range will be retained within each local neighbourhood. However,
2360	in some cases it may be desirable to combine the Dynamic Audio Normalizer's
2361	normalization algorithm with a more "traditional" compression.
2362	For this purpose, the Dynamic Audio Normalizer provides an optional compression
2363	(thresholding) function. If (and only if) the compression feature is enabled,
2364	all input frames will be processed by a soft knee thresholding function prior
2365	to the actual normalization process. Put simply, the thresholding function is
2366	going to prune all samples whose magnitude exceeds a certain threshold value.
2367	However, the Dynamic Audio Normalizer does not simply apply a fixed threshold
2368	value. Instead, the threshold value will be adjusted for each individual
2369	frame.
2370	In general, smaller parameters result in stronger compression, and vice versa.
2371	Values below 3.0 are not recommended, because audible distortion may appear.
2372	@end table
2373
2374	@section earwax
2375
2376	Make audio easier to listen to on headphones.
2377
2378	This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
2379	so that when listened to on headphones the stereo image is moved from
2380	inside your head (standard for headphones) to outside and in front of
2381	the listener (standard for speakers).
2382
2383	Ported from SoX.
2384
2385	@section equalizer
2386
2387	Apply a two-pole peaking equalisation (EQ) filter. With this
2388	filter, the signal-level at and around a selected frequency can
2389	be increased or decreased, whilst (unlike bandpass and bandreject
2390	filters) that at all other frequencies is unchanged.
2391
2392	In order to produce complex equalisation curves, this filter can
2393	be given several times, each with a different central frequency.
2394
2395	The filter accepts the following options:
2396
2397	@table @option
2398	@item frequency, f
2399	Set the filter's central frequency in Hz.
2400
2401	@item width_type
2402	Set method to specify band-width of filter.
2403	@table @option
2404	@item h
2405	Hz
2406	@item q
2407	Q-Factor
2408	@item o
2409	octave
2410	@item s
2411	slope
2412	@end table
2413
2414	@item width, w
2415	Specify the band-width of a filter in width_type units.
2416
2417	@item gain, g
2418	Set the required gain or attenuation in dB.
2419	Beware of clipping when using a positive gain.
2420	@end table
2421
2422	@subsection Examples
2423	@itemize
2424	@item
2425	Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
2426	@example
2427	equalizer=f=1000:width_type=h:width=200:g=-10
2428	@end example
2429
2430	@item
2431	Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
2432	@example
2433	equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
2434	@end example
2435	@end itemize
2436
2437	@section extrastereo
2438
2439	Linearly increases the difference between left and right channels which
2440	adds some sort of "live" effect to playback.
2441
2442	The filter accepts the following options:
2443
2444	@table @option
2445	@item m
2446	Sets the difference coefficient (default: 2.5). 0.0 means mono sound
2447	(average of both channels), with 1.0 sound will be unchanged, with
2448	-1.0 left and right channels will be swapped.
2449
2450	@item c
2451	Enable clipping. By default is enabled.
2452	@end table
2453
2454	@section firequalizer
2455	Apply FIR Equalization using arbitrary frequency response.
2456
2457	The filter accepts the following option:
2458
2459	@table @option
2460	@item gain
2461	Set gain curve equation (in dB). The expression can contain variables:
2462	@table @option
2463	@item f
2464	the evaluated frequency
2465	@item sr
2466	sample rate
2467	@item ch
2468	channel number, set to 0 when multichannels evaluation is disabled
2469	@item chid
2470	channel id, see libavutil/channel_layout.h, set to the first channel id when
2471	multichannels evaluation is disabled
2472	@item chs
2473	number of channels
2474	@item chlayout
2475	channel_layout, see libavutil/channel_layout.h
2476
2477	@end table
2478	and functions:
2479	@table @option
2480	@item gain_interpolate(f)
2481	interpolate gain on frequency f based on gain_entry
2482	@item cubic_interpolate(f)
2483	same as gain_interpolate, but smoother
2484	@end table
2485	This option is also available as command. Default is @code{gain_interpolate(f)}.
2486
2487	@item gain_entry
2488	Set gain entry for gain_interpolate function. The expression can
2489	contain functions:
2490	@table @option
2491	@item entry(f, g)
2492	store gain entry at frequency f with value g
2493	@end table
2494	This option is also available as command.
2495
2496	@item delay
2497	Set filter delay in seconds. Higher value means more accurate.
2498	Default is @code{0.01}.
2499
2500	@item accuracy
2501	Set filter accuracy in Hz. Lower value means more accurate.
2502	Default is @code{5}.
2503
2504	@item wfunc
2505	Set window function. Acceptable values are:
2506	@table @option
2507	@item rectangular
2508	rectangular window, useful when gain curve is already smooth
2509	@item hann
2510	hann window (default)
2511	@item hamming
2512	hamming window
2513	@item blackman
2514	blackman window
2515	@item nuttall3
2516	3-terms continuous 1st derivative nuttall window
2517	@item mnuttall3
2518	minimum 3-terms discontinuous nuttall window
2519	@item nuttall
2520	4-terms continuous 1st derivative nuttall window
2521	@item bnuttall
2522	minimum 4-terms discontinuous nuttall (blackman-nuttall) window
2523	@item bharris
2524	blackman-harris window
2525	@item tukey
2526	tukey window
2527	@end table
2528
2529	@item fixed
2530	If enabled, use fixed number of audio samples. This improves speed when
2531	filtering with large delay. Default is disabled.
2532
2533	@item multi
2534	Enable multichannels evaluation on gain. Default is disabled.
2535
2536	@item zero_phase
2537	Enable zero phase mode by subtracting timestamp to compensate delay.
2538	Default is disabled.
2539
2540	@item scale
2541	Set scale used by gain. Acceptable values are:
2542	@table @option
2543	@item linlin
2544	linear frequency, linear gain
2545	@item linlog
2546	linear frequency, logarithmic (in dB) gain (default)
2547	@item loglin
2548	logarithmic (in octave scale where 20 Hz is 0) frequency, linear gain
2549	@item loglog
2550	logarithmic frequency, logarithmic gain
2551	@end table
2552
2553	@item dumpfile
2554	Set file for dumping, suitable for gnuplot.
2555
2556	@item dumpscale
2557	Set scale for dumpfile. Acceptable values are same with scale option.
2558	Default is linlog.
2559
2560	@item fft2
2561	Enable 2-channel convolution using complex FFT. This improves speed significantly.
2562	Default is disabled.
2563	@end table
2564
2565	@subsection Examples
2566	@itemize
2567	@item
2568	lowpass at 1000 Hz:
2569	@example
2570	firequalizer=gain='if(lt(f,1000), 0, -INF)'
2571	@end example
2572	@item
2573	lowpass at 1000 Hz with gain_entry:
2574	@example
2575	firequalizer=gain_entry='entry(1000,0); entry(1001, -INF)'
2576	@end example
2577	@item
2578	custom equalization:
2579	@example
2580	firequalizer=gain_entry='entry(100,0); entry(400, -4); entry(1000, -6); entry(2000, 0)'
2581	@end example
2582	@item
2583	higher delay with zero phase to compensate delay:
2584	@example
2585	firequalizer=delay=0.1:fixed=on:zero_phase=on
2586	@end example
2587	@item
2588	lowpass on left channel, highpass on right channel:
2589	@example
2590	firequalizer=gain='if(eq(chid,1), gain_interpolate(f), if(eq(chid,2), gain_interpolate(1e6+f), 0))'
2591	:gain_entry='entry(1000, 0); entry(1001,-INF); entry(1e6+1000,0)':multi=on
2592	@end example
2593	@end itemize
2594
2595	@section flanger
2596	Apply a flanging effect to the audio.
2597
2598	The filter accepts the following options:
2599
2600	@table @option
2601	@item delay
2602	Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
2603
2604	@item depth
2605	Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
2606
2607	@item regen
2608	Set percentage regeneration (delayed signal feedback). Range from -95 to 95.
2609	Default value is 0.
2610
2611	@item width
2612	Set percentage of delayed signal mixed with original. Range from 0 to 100.
2613	Default value is 71.
2614
2615	@item speed
2616	Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
2617
2618	@item shape
2619	Set swept wave shape, can be @var{triangular} or @var{sinusoidal}.
2620	Default value is @var{sinusoidal}.
2621
2622	@item phase
2623	Set swept wave percentage-shift for multi channel. Range from 0 to 100.
2624	Default value is 25.
2625
2626	@item interp
2627	Set delay-line interpolation, @var{linear} or @var{quadratic}.
2628	Default is @var{linear}.
2629	@end table
2630
2631	@section hdcd
2632
2633	Decodes High Definition Compatible Digital (HDCD) data. A 16-bit PCM stream with
2634	embedded HDCD codes is expanded into a 20-bit PCM stream.
2635
2636	The filter supports the Peak Extend and Low-level Gain Adjustment features
2637	of HDCD, and detects the Transient Filter flag.
2638
2639	@example
2640	ffmpeg -i HDCD16.flac -af hdcd OUT24.flac
2641	@end example
2642
2643	When using the filter with wav, note the default encoding for wav is 16-bit,
2644	so the resulting 20-bit stream will be truncated back to 16-bit. Use something
2645	like @command{-acodec pcm_s24le} after the filter to get 24-bit PCM output.
2646	@example
2647	ffmpeg -i HDCD16.wav -af hdcd OUT16.wav
2648	ffmpeg -i HDCD16.wav -af hdcd -acodec pcm_s24le OUT24.wav
2649	@end example
2650
2651	The filter accepts the following options:
2652
2653	@table @option
2654	@item disable_autoconvert
2655	Disable any automatic format conversion or resampling in the filter graph.
2656
2657	@item process_stereo
2658	Process the stereo channels together. If target_gain does not match between
2659	channels, consider it invalid and use the last valid target_gain.
2660
2661	@item cdt_ms
2662	Set the code detect timer period in ms.
2663
2664	@item force_pe
2665	Always extend peaks above -3dBFS even if PE isn't signaled.
2666
2667	@item analyze_mode
2668	Replace audio with a solid tone and adjust the amplitude to signal some
2669	specific aspect of the decoding process. The output file can be loaded in
2670	an audio editor alongside the original to aid analysis.
2671
2672	@code{analyze_mode=pe:force_pe=true} can be used to see all samples above the PE level.
2673
2674	Modes are:
2675	@table @samp
2676	@item 0, off
2677	Disabled
2678	@item 1, lle
2679	Gain adjustment level at each sample
2680	@item 2, pe
2681	Samples where peak extend occurs
2682	@item 3, cdt
2683	Samples where the code detect timer is active
2684	@item 4, tgm
2685	Samples where the target gain does not match between channels
2686	@end table
2687	@end table
2688
2689	@section highpass
2690
2691	Apply a high-pass filter with 3dB point frequency.
2692	The filter can be either single-pole, or double-pole (the default).
2693	The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2694
2695	The filter accepts the following options:
2696
2697	@table @option
2698	@item frequency, f
2699	Set frequency in Hz. Default is 3000.
2700
2701	@item poles, p
2702	Set number of poles. Default is 2.
2703
2704	@item width_type
2705	Set method to specify band-width of filter.
2706	@table @option
2707	@item h
2708	Hz
2709	@item q
2710	Q-Factor
2711	@item o
2712	octave
2713	@item s
2714	slope
2715	@end table
2716
2717	@item width, w
2718	Specify the band-width of a filter in width_type units.
2719	Applies only to double-pole filter.
2720	The default is 0.707q and gives a Butterworth response.
2721	@end table
2722
2723	@section join
2724
2725	Join multiple input streams into one multi-channel stream.
2726
2727	It accepts the following parameters:
2728	@table @option
2729
2730	@item inputs
2731	The number of input streams. It defaults to 2.
2732
2733	@item channel_layout
2734	The desired output channel layout. It defaults to stereo.
2735
2736	@item map
2737	Map channels from inputs to output. The argument is a '|'-separated list of
2738	mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
2739	form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
2740	can be either the name of the input channel (e.g. FL for front left) or its
2741	index in the specified input stream. @var{out_channel} is the name of the output
2742	channel.
2743	@end table
2744
2745	The filter will attempt to guess the mappings when they are not specified
2746	explicitly. It does so by first trying to find an unused matching input channel
2747	and if that fails it picks the first unused input channel.
2748
2749	Join 3 inputs (with properly set channel layouts):
2750	@example
2751	ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
2752	@end example
2753
2754	Build a 5.1 output from 6 single-channel streams:
2755	@example
2756	ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
2757	'join=inputs=6:channel_layout=5.1:map=0.0-FL|1.0-FR|2.0-FC|3.0-SL|4.0-SR|5.0-LFE'
2758	out
2759	@end example
2760
2761	@section ladspa
2762
2763	Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
2764
2765	To enable compilation of this filter you need to configure FFmpeg with
2766	@code{--enable-ladspa}.
2767
2768	@table @option
2769	@item file, f
2770	Specifies the name of LADSPA plugin library to load. If the environment
2771	variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
2772	each one of the directories specified by the colon separated list in
2773	@env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
2774	this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
2775	@file{/usr/lib/ladspa/}.
2776
2777	@item plugin, p
2778	Specifies the plugin within the library. Some libraries contain only
2779	one plugin, but others contain many of them. If this is not set filter
2780	will list all available plugins within the specified library.
2781
2782	@item controls, c
2783	Set the '|' separated list of controls which are zero or more floating point
2784	values that determine the behavior of the loaded plugin (for example delay,
2785	threshold or gain).
2786	Controls need to be defined using the following syntax:
2787	c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
2788	@var{valuei} is the value set on the @var{i}-th control.
2789	Alternatively they can be also defined using the following syntax:
2790	@var{value0}|@var{value1}|@var{value2}|..., where
2791	@var{valuei} is the value set on the @var{i}-th control.
2792	If @option{controls} is set to @code{help}, all available controls and
2793	their valid ranges are printed.
2794
2795	@item sample_rate, s
2796	Specify the sample rate, default to 44100. Only used if plugin have
2797	zero inputs.
2798
2799	@item nb_samples, n
2800	Set the number of samples per channel per each output frame, default
2801	is 1024. Only used if plugin have zero inputs.
2802
2803	@item duration, d
2804	Set the minimum duration of the sourced audio. See
2805	@ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
2806	for the accepted syntax.
2807	Note that the resulting duration may be greater than the specified duration,
2808	as the generated audio is always cut at the end of a complete frame.
2809	If not specified, or the expressed duration is negative, the audio is
2810	supposed to be generated forever.
2811	Only used if plugin have zero inputs.
2812
2813	@end table
2814
2815	@subsection Examples
2816
2817	@itemize
2818	@item
2819	List all available plugins within amp (LADSPA example plugin) library:
2820	@example
2821	ladspa=file=amp
2822	@end example
2823
2824	@item
2825	List all available controls and their valid ranges for @code{vcf_notch}
2826	plugin from @code{VCF} library:
2827	@example
2828	ladspa=f=vcf:p=vcf_notch:c=help
2829	@end example
2830
2831	@item
2832	Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
2833	plugin library:
2834	@example
2835	ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
2836	@end example
2837
2838	@item
2839	Add reverberation to the audio using TAP-plugins
2840	(Tom's Audio Processing plugins):
2841	@example
2842	ladspa=file=tap_reverb:tap_reverb
2843	@end example
2844
2845	@item
2846	Generate white noise, with 0.2 amplitude:
2847	@example
2848	ladspa=file=cmt:noise_source_white:c=c0=.2
2849	@end example
2850
2851	@item
2852	Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
2853	@code{C* Audio Plugin Suite} (CAPS) library:
2854	@example
2855	ladspa=file=caps:Click:c=c1=20'
2856	@end example
2857
2858	@item
2859	Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
2860	@example
2861	ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
2862	@end example
2863
2864	@item
2865	Increase volume by 20dB using fast lookahead limiter from Steve Harris
2866	@code{SWH Plugins} collection:
2867	@example
2868	ladspa=fast_lookahead_limiter_1913:fastLookaheadLimiter:20|0|2
2869	@end example
2870
2871	@item
2872	Attenuate low frequencies using Multiband EQ from Steve Harris
2873	@code{SWH Plugins} collection:
2874	@example
2875	ladspa=mbeq_1197:mbeq:-24|-24|-24|0|0|0|0|0|0|0|0|0|0|0|0
2876	@end example
2877	@end itemize
2878
2879	@subsection Commands
2880
2881	This filter supports the following commands:
2882	@table @option
2883	@item cN
2884	Modify the @var{N}-th control value.
2885
2886	If the specified value is not valid, it is ignored and prior one is kept.
2887	@end table
2888
2889	@section loudnorm
2890
2891	EBU R128 loudness normalization. Includes both dynamic and linear normalization modes.
2892	Support for both single pass (livestreams, files) and double pass (files) modes.
2893	This algorithm can target IL, LRA, and maximum true peak.
2894
2895	The filter accepts the following options:
2896
2897	@table @option
2898	@item I, i
2899	Set integrated loudness target.
2900	Range is -70.0 - -5.0. Default value is -24.0.
2901
2902	@item LRA, lra
2903	Set loudness range target.
2904	Range is 1.0 - 20.0. Default value is 7.0.
2905
2906	@item TP, tp
2907	Set maximum true peak.
2908	Range is -9.0 - +0.0. Default value is -2.0.
2909
2910	@item measured_I, measured_i
2911	Measured IL of input file.
2912	Range is -99.0 - +0.0.
2913
2914	@item measured_LRA, measured_lra
2915	Measured LRA of input file.
2916	Range is 0.0 - 99.0.
2917
2918	@item measured_TP, measured_tp
2919	Measured true peak of input file.
2920	Range is -99.0 - +99.0.
2921
2922	@item measured_thresh
2923	Measured threshold of input file.
2924	Range is -99.0 - +0.0.
2925
2926	@item offset
2927	Set offset gain. Gain is applied before the true-peak limiter.
2928	Range is -99.0 - +99.0. Default is +0.0.
2929
2930	@item linear
2931	Normalize linearly if possible.
2932	measured_I, measured_LRA, measured_TP, and measured_thresh must also
2933	to be specified in order to use this mode.
2934	Options are true or false. Default is true.
2935
2936	@item dual_mono
2937	Treat mono input files as "dual-mono". If a mono file is intended for playback
2938	on a stereo system, its EBU R128 measurement will be perceptually incorrect.
2939	If set to @code{true}, this option will compensate for this effect.
2940	Multi-channel input files are not affected by this option.
2941	Options are true or false. Default is false.
2942
2943	@item print_format
2944	Set print format for stats. Options are summary, json, or none.
2945	Default value is none.
2946	@end table
2947
2948	@section lowpass
2949
2950	Apply a low-pass filter with 3dB point frequency.
2951	The filter can be either single-pole or double-pole (the default).
2952	The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2953
2954	The filter accepts the following options:
2955
2956	@table @option
2957	@item frequency, f
2958	Set frequency in Hz. Default is 500.
2959
2960	@item poles, p
2961	Set number of poles. Default is 2.
2962
2963	@item width_type
2964	Set method to specify band-width of filter.
2965	@table @option
2966	@item h
2967	Hz
2968	@item q
2969	Q-Factor
2970	@item o
2971	octave
2972	@item s
2973	slope
2974	@end table
2975
2976	@item width, w
2977	Specify the band-width of a filter in width_type units.
2978	Applies only to double-pole filter.
2979	The default is 0.707q and gives a Butterworth response.
2980	@end table
2981
2982	@anchor{pan}
2983	@section pan
2984
2985	Mix channels with specific gain levels. The filter accepts the output
2986	channel layout followed by a set of channels definitions.
2987
2988	This filter is also designed to efficiently remap the channels of an audio
2989	stream.
2990
2991	The filter accepts parameters of the form:
2992	"@var{l}|@var{outdef}|@var{outdef}|..."
2993
2994	@table @option
2995	@item l
2996	output channel layout or number of channels
2997
2998	@item outdef
2999	output channel specification, of the form:
3000	"@var{out_name}=[@var{gain}*]@var{in_name}[(+-)[@var{gain}*]@var{in_name}...]"
3001
3002	@item out_name
3003	output channel to define, either a channel name (FL, FR, etc.) or a channel
3004	number (c0, c1, etc.)
3005
3006	@item gain
3007	multiplicative coefficient for the channel, 1 leaving the volume unchanged
3008
3009	@item in_name
3010	input channel to use, see out_name for details; it is not possible to mix
3011	named and numbered input channels
3012	@end table
3013
3014	If the `=' in a channel specification is replaced by `<', then the gains for
3015	that specification will be renormalized so that the total is 1, thus
3016	avoiding clipping noise.
3017
3018	@subsection Mixing examples
3019
3020	For example, if you want to down-mix from stereo to mono, but with a bigger
3021	factor for the left channel:
3022	@example
3023	pan=1c|c0=0.9*c0+0.1*c1
3024	@end example
3025
3026	A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
3027	7-channels surround:
3028	@example
3029	pan=stereo| FL < FL + 0.5*FC + 0.6*BL + 0.6*SL | FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
3030	@end example
3031
3032	Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
3033	that should be preferred (see "-ac" option) unless you have very specific
3034	needs.
3035
3036	@subsection Remapping examples
3037
3038	The channel remapping will be effective if, and only if:
3039
3040	@itemize
3041	@item gain coefficients are zeroes or ones,
3042	@item only one input per channel output,
3043	@end itemize
3044
3045	If all these conditions are satisfied, the filter will notify the user ("Pure
3046	channel mapping detected"), and use an optimized and lossless method to do the
3047	remapping.
3048
3049	For example, if you have a 5.1 source and want a stereo audio stream by
3050	dropping the extra channels:
3051	@example
3052	pan="stereo| c0=FL | c1=FR"
3053	@end example
3054
3055	Given the same source, you can also switch front left and front right channels
3056	and keep the input channel layout:
3057	@example
3058	pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
3059	@end example
3060
3061	If the input is a stereo audio stream, you can mute the front left channel (and
3062	still keep the stereo channel layout) with:
3063	@example
3064	pan="stereo|c1=c1"
3065	@end example
3066
3067	Still with a stereo audio stream input, you can copy the right channel in both
3068	front left and right:
3069	@example
3070	pan="stereo| c0=FR | c1=FR"
3071	@end example
3072
3073	@section replaygain
3074
3075	ReplayGain scanner filter. This filter takes an audio stream as an input and
3076	outputs it unchanged.
3077	At end of filtering it displays @code{track_gain} and @code{track_peak}.
3078
3079	@section resample
3080
3081	Convert the audio sample format, sample rate and channel layout. It is
3082	not meant to be used directly.
3083
3084	@section rubberband
3085	Apply time-stretching and pitch-shifting with librubberband.
3086
3087	The filter accepts the following options:
3088
3089	@table @option
3090	@item tempo
3091	Set tempo scale factor.
3092
3093	@item pitch
3094	Set pitch scale factor.
3095
3096	@item transients
3097	Set transients detector.
3098	Possible values are:
3099	@table @var
3100	@item crisp
3101	@item mixed
3102	@item smooth
3103	@end table
3104
3105	@item detector
3106	Set detector.
3107	Possible values are:
3108	@table @var
3109	@item compound
3110	@item percussive
3111	@item soft
3112	@end table
3113
3114	@item phase
3115	Set phase.
3116	Possible values are:
3117	@table @var
3118	@item laminar
3119	@item independent
3120	@end table
3121
3122	@item window
3123	Set processing window size.
3124	Possible values are:
3125	@table @var
3126	@item standard
3127	@item short
3128	@item long
3129	@end table
3130
3131	@item smoothing
3132	Set smoothing.
3133	Possible values are:
3134	@table @var
3135	@item off
3136	@item on
3137	@end table
3138
3139	@item formant
3140	Enable formant preservation when shift pitching.
3141	Possible values are:
3142	@table @var
3143	@item shifted
3144	@item preserved
3145	@end table
3146
3147	@item pitchq
3148	Set pitch quality.
3149	Possible values are:
3150	@table @var
3151	@item quality
3152	@item speed
3153	@item consistency
3154	@end table
3155
3156	@item channels
3157	Set channels.
3158	Possible values are:
3159	@table @var
3160	@item apart
3161	@item together
3162	@end table
3163	@end table
3164
3165	@section sidechaincompress
3166
3167	This filter acts like normal compressor but has the ability to compress
3168	detected signal using second input signal.
3169	It needs two input streams and returns one output stream.
3170	First input stream will be processed depending on second stream signal.
3171	The filtered signal then can be filtered with other filters in later stages of
3172	processing. See @ref{pan} and @ref{amerge} filter.
3173
3174	The filter accepts the following options:
3175
3176	@table @option
3177	@item level_in
3178	Set input gain. Default is 1. Range is between 0.015625 and 64.
3179
3180	@item threshold
3181	If a signal of second stream raises above this level it will affect the gain
3182	reduction of first stream.
3183	By default is 0.125. Range is between 0.00097563 and 1.
3184
3185	@item ratio
3186	Set a ratio about which the signal is reduced. 1:2 means that if the level
3187	raised 4dB above the threshold, it will be only 2dB above after the reduction.
3188	Default is 2. Range is between 1 and 20.
3189
3190	@item attack
3191	Amount of milliseconds the signal has to rise above the threshold before gain
3192	reduction starts. Default is 20. Range is between 0.01 and 2000.
3193
3194	@item release
3195	Amount of milliseconds the signal has to fall below the threshold before
3196	reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
3197
3198	@item makeup
3199	Set the amount by how much signal will be amplified after processing.
3200	Default is 2. Range is from 1 and 64.
3201
3202	@item knee
3203	Curve the sharp knee around the threshold to enter gain reduction more softly.
3204	Default is 2.82843. Range is between 1 and 8.
3205
3206	@item link
3207	Choose if the @code{average} level between all channels of side-chain stream
3208	or the louder(@code{maximum}) channel of side-chain stream affects the
3209	reduction. Default is @code{average}.
3210
3211	@item detection
3212	Should the exact signal be taken in case of @code{peak} or an RMS one in case
3213	of @code{rms}. Default is @code{rms} which is mainly smoother.
3214
3215	@item level_sc
3216	Set sidechain gain. Default is 1. Range is between 0.015625 and 64.
3217
3218	@item mix
3219	How much to use compressed signal in output. Default is 1.
3220	Range is between 0 and 1.
3221	@end table
3222
3223	@subsection Examples
3224
3225	@itemize
3226	@item
3227	Full ffmpeg example taking 2 audio inputs, 1st input to be compressed
3228	depending on the signal of 2nd input and later compressed signal to be
3229	merged with 2nd input:
3230	@example
3231	ffmpeg -i main.flac -i sidechain.flac -filter_complex "[1:a]asplit=2[sc][mix];[0:a][sc]sidechaincompress[compr];[compr][mix]amerge"
3232	@end example
3233	@end itemize
3234
3235	@section sidechaingate
3236
3237	A sidechain gate acts like a normal (wideband) gate but has the ability to
3238	filter the detected signal before sending it to the gain reduction stage.
3239	Normally a gate uses the full range signal to detect a level above the
3240	threshold.
3241	For example: If you cut all lower frequencies from your sidechain signal
3242	the gate will decrease the volume of your track only if not enough highs
3243	appear. With this technique you are able to reduce the resonation of a
3244	natural drum or remove "rumbling" of muted strokes from a heavily distorted
3245	guitar.
3246	It needs two input streams and returns one output stream.
3247	First input stream will be processed depending on second stream signal.
3248
3249	The filter accepts the following options:
3250
3251	@table @option
3252	@item level_in
3253	Set input level before filtering.
3254	Default is 1. Allowed range is from 0.015625 to 64.
3255
3256	@item range
3257	Set the level of gain reduction when the signal is below the threshold.
3258	Default is 0.06125. Allowed range is from 0 to 1.
3259
3260	@item threshold
3261	If a signal rises above this level the gain reduction is released.
3262	Default is 0.125. Allowed range is from 0 to 1.
3263
3264	@item ratio
3265	Set a ratio about which the signal is reduced.
3266	Default is 2. Allowed range is from 1 to 9000.
3267
3268	@item attack
3269	Amount of milliseconds the signal has to rise above the threshold before gain
3270	reduction stops.
3271	Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
3272
3273	@item release
3274	Amount of milliseconds the signal has to fall below the threshold before the
3275	reduction is increased again. Default is 250 milliseconds.
3276	Allowed range is from 0.01 to 9000.
3277
3278	@item makeup
3279	Set amount of amplification of signal after processing.
3280	Default is 1. Allowed range is from 1 to 64.
3281
3282	@item knee
3283	Curve the sharp knee around the threshold to enter gain reduction more softly.
3284	Default is 2.828427125. Allowed range is from 1 to 8.
3285
3286	@item detection
3287	Choose if exact signal should be taken for detection or an RMS like one.
3288	Default is rms. Can be peak or rms.
3289
3290	@item link
3291	Choose if the average level between all channels or the louder channel affects
3292	the reduction.
3293	Default is average. Can be average or maximum.
3294
3295	@item level_sc
3296	Set sidechain gain. Default is 1. Range is from 0.015625 to 64.
3297	@end table
3298
3299	@section silencedetect
3300
3301	Detect silence in an audio stream.
3302
3303	This filter logs a message when it detects that the input audio volume is less
3304	or equal to a noise tolerance value for a duration greater or equal to the
3305	minimum detected noise duration.
3306
3307	The printed times and duration are expressed in seconds.
3308
3309	The filter accepts the following options:
3310
3311	@table @option
3312	@item duration, d
3313	Set silence duration until notification (default is 2 seconds).
3314
3315	@item noise, n
3316	Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
3317	specified value) or amplitude ratio. Default is -60dB, or 0.001.
3318	@end table
3319
3320	@subsection Examples
3321
3322	@itemize
3323	@item
3324	Detect 5 seconds of silence with -50dB noise tolerance:
3325	@example
3326	silencedetect=n=-50dB:d=5
3327	@end example
3328
3329	@item
3330	Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
3331	tolerance in @file{silence.mp3}:
3332	@example
3333	ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
3334	@end example
3335	@end itemize
3336
3337	@section silenceremove
3338
3339	Remove silence from the beginning, middle or end of the audio.
3340
3341	The filter accepts the following options:
3342
3343	@table @option
3344	@item start_periods
3345	This value is used to indicate if audio should be trimmed at beginning of
3346	the audio. A value of zero indicates no silence should be trimmed from the
3347	beginning. When specifying a non-zero value, it trims audio up until it
3348	finds non-silence. Normally, when trimming silence from beginning of audio
3349	the @var{start_periods} will be @code{1} but it can be increased to higher
3350	values to trim all audio up to specific count of non-silence periods.
3351	Default value is @code{0}.
3352
3353	@item start_duration
3354	Specify the amount of time that non-silence must be detected before it stops
3355	trimming audio. By increasing the duration, bursts of noises can be treated
3356	as silence and trimmed off. Default value is @code{0}.
3357
3358	@item start_threshold
3359	This indicates what sample value should be treated as silence. For digital
3360	audio, a value of @code{0} may be fine but for audio recorded from analog,
3361	you may wish to increase the value to account for background noise.
3362	Can be specified in dB (in case "dB" is appended to the specified value)
3363	or amplitude ratio. Default value is @code{0}.
3364
3365	@item stop_periods
3366	Set the count for trimming silence from the end of audio.
3367	To remove silence from the middle of a file, specify a @var{stop_periods}
3368	that is negative. This value is then treated as a positive value and is
3369	used to indicate the effect should restart processing as specified by
3370	@var{start_periods}, making it suitable for removing periods of silence
3371	in the middle of the audio.
3372	Default value is @code{0}.
3373
3374	@item stop_duration
3375	Specify a duration of silence that must exist before audio is not copied any
3376	more. By specifying a higher duration, silence that is wanted can be left in
3377	the audio.
3378	Default value is @code{0}.
3379
3380	@item stop_threshold
3381	This is the same as @option{start_threshold} but for trimming silence from
3382	the end of audio.
3383	Can be specified in dB (in case "dB" is appended to the specified value)
3384	or amplitude ratio. Default value is @code{0}.
3385
3386	@item leave_silence
3387	This indicates that @var{stop_duration} length of audio should be left intact
3388	at the beginning of each period of silence.
3389	For example, if you want to remove long pauses between words but do not want
3390	to remove the pauses completely. Default value is @code{0}.
3391
3392	@item detection
3393	Set how is silence detected. Can be @code{rms} or @code{peak}. Second is faster
3394	and works better with digital silence which is exactly 0.
3395	Default value is @code{rms}.
3396
3397	@item window
3398	Set ratio used to calculate size of window for detecting silence.
3399	Default value is @code{0.02}. Allowed range is from @code{0} to @code{10}.
3400	@end table
3401
3402	@subsection Examples
3403
3404	@itemize
3405	@item
3406	The following example shows how this filter can be used to start a recording
3407	that does not contain the delay at the start which usually occurs between
3408	pressing the record button and the start of the performance:
3409	@example
3410	silenceremove=1:5:0.02
3411	@end example
3412
3413	@item
3414	Trim all silence encountered from beginning to end where there is more than 1
3415	second of silence in audio:
3416	@example
3417	silenceremove=0:0:0:-1:1:-90dB
3418	@end example
3419	@end itemize
3420
3421	@section sofalizer
3422
3423	SOFAlizer uses head-related transfer functions (HRTFs) to create virtual
3424	loudspeakers around the user for binaural listening via headphones (audio
3425	formats up to 9 channels supported).
3426	The HRTFs are stored in SOFA files (see @url{http://www.sofacoustics.org/} for a database).
3427	SOFAlizer is developed at the Acoustics Research Institute (ARI) of the
3428	Austrian Academy of Sciences.
3429
3430	To enable compilation of this filter you need to configure FFmpeg with
3431	@code{--enable-netcdf}.
3432
3433	The filter accepts the following options:
3434
3435	@table @option
3436	@item sofa
3437	Set the SOFA file used for rendering.
3438
3439	@item gain
3440	Set gain applied to audio. Value is in dB. Default is 0.
3441
3442	@item rotation
3443	Set rotation of virtual loudspeakers in deg. Default is 0.
3444
3445	@item elevation
3446	Set elevation of virtual speakers in deg. Default is 0.
3447
3448	@item radius
3449	Set distance in meters between loudspeakers and the listener with near-field
3450	HRTFs. Default is 1.
3451
3452	@item type
3453	Set processing type. Can be @var{time} or @var{freq}. @var{time} is
3454	processing audio in time domain which is slow.
3455	@var{freq} is processing audio in frequency domain which is fast.
3456	Default is @var{freq}.
3457
3458	@item speakers
3459	Set custom positions of virtual loudspeakers. Syntax for this option is:
3460	<CH> <AZIM> <ELEV>[|<CH> <AZIM> <ELEV>|...].
3461	Each virtual loudspeaker is described with short channel name following with
3462	azimuth and elevation in degreees.
3463	Each virtual loudspeaker description is separated by '|'.
3464	For example to override front left and front right channel positions use:
3465	'speakers=FL 45 15|FR 345 15'.
3466	Descriptions with unrecognised channel names are ignored.
3467	@end table
3468
3469	@subsection Examples
3470
3471	@itemize
3472	@item
3473	Using ClubFritz6 sofa file:
3474	@example
3475	sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=1
3476	@end example
3477
3478	@item
3479	Using ClubFritz12 sofa file and bigger radius with small rotation:
3480	@example
3481	sofalizer=sofa=/path/to/ClubFritz12.sofa:type=freq:radius=2:rotation=5
3482	@end example
3483
3484	@item
3485	Similar as above but with custom speaker positions for front left, front right, back left and back right
3486	and also with custom gain:
3487	@example
3488	"sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=2:speakers=FL 45|FR 315|BL 135|BR 225:gain=28"
3489	@end example
3490	@end itemize
3491
3492	@section stereotools
3493
3494	This filter has some handy utilities to manage stereo signals, for converting
3495	M/S stereo recordings to L/R signal while having control over the parameters
3496	or spreading the stereo image of master track.
3497
3498	The filter accepts the following options:
3499
3500	@table @option
3501	@item level_in
3502	Set input level before filtering for both channels. Defaults is 1.
3503	Allowed range is from 0.015625 to 64.
3504
3505	@item level_out
3506	Set output level after filtering for both channels. Defaults is 1.
3507	Allowed range is from 0.015625 to 64.
3508
3509	@item balance_in
3510	Set input balance between both channels. Default is 0.
3511	Allowed range is from -1 to 1.
3512
3513	@item balance_out
3514	Set output balance between both channels. Default is 0.
3515	Allowed range is from -1 to 1.
3516
3517	@item softclip
3518	Enable softclipping. Results in analog distortion instead of harsh digital 0dB
3519	clipping. Disabled by default.
3520
3521	@item mutel
3522	Mute the left channel. Disabled by default.
3523
3524	@item muter
3525	Mute the right channel. Disabled by default.
3526
3527	@item phasel
3528	Change the phase of the left channel. Disabled by default.
3529
3530	@item phaser
3531	Change the phase of the right channel. Disabled by default.
3532
3533	@item mode
3534	Set stereo mode. Available values are:
3535
3536	@table @samp
3537	@item lr>lr
3538	Left/Right to Left/Right, this is default.
3539
3540	@item lr>ms
3541	Left/Right to Mid/Side.
3542
3543	@item ms>lr
3544	Mid/Side to Left/Right.
3545
3546	@item lr>ll
3547	Left/Right to Left/Left.
3548
3549	@item lr>rr
3550	Left/Right to Right/Right.
3551
3552	@item lr>l+r
3553	Left/Right to Left + Right.
3554
3555	@item lr>rl
3556	Left/Right to Right/Left.
3557	@end table
3558
3559	@item slev
3560	Set level of side signal. Default is 1.
3561	Allowed range is from 0.015625 to 64.
3562
3563	@item sbal
3564	Set balance of side signal. Default is 0.
3565	Allowed range is from -1 to 1.
3566
3567	@item mlev
3568	Set level of the middle signal. Default is 1.
3569	Allowed range is from 0.015625 to 64.
3570
3571	@item mpan
3572	Set middle signal pan. Default is 0. Allowed range is from -1 to 1.
3573
3574	@item base
3575	Set stereo base between mono and inversed channels. Default is 0.
3576	Allowed range is from -1 to 1.
3577
3578	@item delay
3579	Set delay in milliseconds how much to delay left from right channel and
3580	vice versa. Default is 0. Allowed range is from -20 to 20.
3581
3582	@item sclevel
3583	Set S/C level. Default is 1. Allowed range is from 1 to 100.
3584
3585	@item phase
3586	Set the stereo phase in degrees. Default is 0. Allowed range is from 0 to 360.
3587	@end table
3588
3589	@subsection Examples
3590
3591	@itemize
3592	@item
3593	Apply karaoke like effect:
3594	@example
3595	stereotools=mlev=0.015625
3596	@end example
3597
3598	@item
3599	Convert M/S signal to L/R:
3600	@example
3601	"stereotools=mode=ms>lr"
3602	@end example
3603	@end itemize
3604
3605	@section stereowiden
3606
3607	This filter enhance the stereo effect by suppressing signal common to both
3608	channels and by delaying the signal of left into right and vice versa,
3609	thereby widening the stereo effect.
3610
3611	The filter accepts the following options:
3612
3613	@table @option
3614	@item delay
3615	Time in milliseconds of the delay of left signal into right and vice versa.
3616	Default is 20 milliseconds.
3617
3618	@item feedback
3619	Amount of gain in delayed signal into right and vice versa. Gives a delay
3620	effect of left signal in right output and vice versa which gives widening
3621	effect. Default is 0.3.
3622
3623	@item crossfeed
3624	Cross feed of left into right with inverted phase. This helps in suppressing
3625	the mono. If the value is 1 it will cancel all the signal common to both
3626	channels. Default is 0.3.
3627
3628	@item drymix
3629	Set level of input signal of original channel. Default is 0.8.
3630	@end table
3631
3632	@section treble
3633
3634	Boost or cut treble (upper) frequencies of the audio using a two-pole
3635	shelving filter with a response similar to that of a standard
3636	hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
3637
3638	The filter accepts the following options:
3639
3640	@table @option
3641	@item gain, g
3642	Give the gain at whichever is the lower of ~22 kHz and the
3643	Nyquist frequency. Its useful range is about -20 (for a large cut)
3644	to +20 (for a large boost). Beware of clipping when using a positive gain.
3645
3646	@item frequency, f
3647	Set the filter's central frequency and so can be used
3648	to extend or reduce the frequency range to be boosted or cut.
3649	The default value is @code{3000} Hz.
3650
3651	@item width_type
3652	Set method to specify band-width of filter.
3653	@table @option
3654	@item h
3655	Hz
3656	@item q
3657	Q-Factor
3658	@item o
3659	octave
3660	@item s
3661	slope
3662	@end table
3663
3664	@item width, w
3665	Determine how steep is the filter's shelf transition.
3666	@end table
3667
3668	@section tremolo
3669
3670	Sinusoidal amplitude modulation.
3671
3672	The filter accepts the following options:
3673
3674	@table @option
3675	@item f
3676	Modulation frequency in Hertz. Modulation frequencies in the subharmonic range
3677	(20 Hz or lower) will result in a tremolo effect.
3678	This filter may also be used as a ring modulator by specifying
3679	a modulation frequency higher than 20 Hz.
3680	Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3681
3682	@item d
3683	Depth of modulation as a percentage. Range is 0.0 - 1.0.
3684	Default value is 0.5.
3685	@end table
3686
3687	@section vibrato
3688
3689	Sinusoidal phase modulation.
3690
3691	The filter accepts the following options:
3692
3693	@table @option
3694	@item f
3695	Modulation frequency in Hertz.
3696	Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3697
3698	@item d
3699	Depth of modulation as a percentage. Range is 0.0 - 1.0.
3700	Default value is 0.5.
3701	@end table
3702
3703	@section volume
3704
3705	Adjust the input audio volume.
3706
3707	It accepts the following parameters:
3708	@table @option
3709
3710	@item volume
3711	Set audio volume expression.
3712
3713	Output values are clipped to the maximum value.
3714
3715	The output audio volume is given by the relation:
3716	@example
3717	@var{output_volume} = @var{volume} * @var{input_volume}
3718	@end example
3719
3720	The default value for @var{volume} is "1.0".
3721
3722	@item precision
3723	This parameter represents the mathematical precision.
3724
3725	It determines which input sample formats will be allowed, which affects the
3726	precision of the volume scaling.
3727
3728	@table @option
3729	@item fixed
3730	8-bit fixed-point; this limits input sample format to U8, S16, and S32.
3731	@item float
3732	32-bit floating-point; this limits input sample format to FLT. (default)
3733	@item double
3734	64-bit floating-point; this limits input sample format to DBL.
3735	@end table
3736
3737	@item replaygain
3738	Choose the behaviour on encountering ReplayGain side data in input frames.
3739
3740	@table @option
3741	@item drop
3742	Remove ReplayGain side data, ignoring its contents (the default).
3743
3744	@item ignore
3745	Ignore ReplayGain side data, but leave it in the frame.
3746
3747	@item track
3748	Prefer the track gain, if present.
3749
3750	@item album
3751	Prefer the album gain, if present.
3752	@end table
3753
3754	@item replaygain_preamp
3755	Pre-amplification gain in dB to apply to the selected replaygain gain.
3756
3757	Default value for @var{replaygain_preamp} is 0.0.
3758
3759	@item eval
3760	Set when the volume expression is evaluated.
3761
3762	It accepts the following values:
3763	@table @samp
3764	@item once
3765	only evaluate expression once during the filter initialization, or
3766	when the @samp{volume} command is sent
3767
3768	@item frame
3769	evaluate expression for each incoming frame
3770	@end table
3771
3772	Default value is @samp{once}.
3773	@end table
3774
3775	The volume expression can contain the following parameters.
3776
3777	@table @option
3778	@item n
3779	frame number (starting at zero)
3780	@item nb_channels
3781	number of channels
3782	@item nb_consumed_samples
3783	number of samples consumed by the filter
3784	@item nb_samples
3785	number of samples in the current frame
3786	@item pos
3787	original frame position in the file
3788	@item pts
3789	frame PTS
3790	@item sample_rate
3791	sample rate
3792	@item startpts
3793	PTS at start of stream
3794	@item startt
3795	time at start of stream
3796	@item t
3797	frame time
3798	@item tb
3799	timestamp timebase
3800	@item volume
3801	last set volume value
3802	@end table
3803
3804	Note that when @option{eval} is set to @samp{once} only the
3805	@var{sample_rate} and @var{tb} variables are available, all other
3806	variables will evaluate to NAN.
3807
3808	@subsection Commands
3809
3810	This filter supports the following commands:
3811	@table @option
3812	@item volume
3813	Modify the volume expression.
3814	The command accepts the same syntax of the corresponding option.
3815
3816	If the specified expression is not valid, it is kept at its current
3817	value.
3818	@item replaygain_noclip
3819	Prevent clipping by limiting the gain applied.
3820
3821	Default value for @var{replaygain_noclip} is 1.
3822
3823	@end table
3824
3825	@subsection Examples
3826
3827	@itemize
3828	@item
3829	Halve the input audio volume:
3830	@example
3831	volume=volume=0.5
3832	volume=volume=1/2
3833	volume=volume=-6.0206dB
3834	@end example
3835
3836	In all the above example the named key for @option{volume} can be
3837	omitted, for example like in:
3838	@example
3839	volume=0.5
3840	@end example
3841
3842	@item
3843	Increase input audio power by 6 decibels using fixed-point precision:
3844	@example
3845	volume=volume=6dB:precision=fixed
3846	@end example
3847
3848	@item
3849	Fade volume after time 10 with an annihilation period of 5 seconds:
3850	@example
3851	volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
3852	@end example
3853	@end itemize
3854
3855	@section volumedetect
3856
3857	Detect the volume of the input video.
3858
3859	The filter has no parameters. The input is not modified. Statistics about
3860	the volume will be printed in the log when the input stream end is reached.
3861
3862	In particular it will show the mean volume (root mean square), maximum
3863	volume (on a per-sample basis), and the beginning of a histogram of the
3864	registered volume values (from the maximum value to a cumulated 1/1000 of
3865	the samples).
3866
3867	All volumes are in decibels relative to the maximum PCM value.
3868
3869	@subsection Examples
3870
3871	Here is an excerpt of the output:
3872	@example
3873	[Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
3874	[Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
3875	[Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
3876	[Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
3877	[Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
3878	[Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
3879	[Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
3880	[Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
3881	[Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
3882	@end example
3883
3884	It means that:
3885	@itemize
3886	@item
3887	The mean square energy is approximately -27 dB, or 10^-2.7.
3888	@item
3889	The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
3890	@item
3891	There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
3892	@end itemize
3893
3894	In other words, raising the volume by +4 dB does not cause any clipping,
3895	raising it by +5 dB causes clipping for 6 samples, etc.
3896
3897	@c man end AUDIO FILTERS
3898
3899	@chapter Audio Sources
3900	@c man begin AUDIO SOURCES
3901
3902	Below is a description of the currently available audio sources.
3903
3904	@section abuffer
3905
3906	Buffer audio frames, and make them available to the filter chain.
3907
3908	This source is mainly intended for a programmatic use, in particular
3909	through the interface defined in @file{libavfilter/asrc_abuffer.h}.
3910
3911	It accepts the following parameters:
3912	@table @option
3913
3914	@item time_base
3915	The timebase which will be used for timestamps of submitted frames. It must be
3916	either a floating-point number or in @var{numerator}/@var{denominator} form.
3917
3918	@item sample_rate
3919	The sample rate of the incoming audio buffers.
3920
3921	@item sample_fmt
3922	The sample format of the incoming audio buffers.
3923	Either a sample format name or its corresponding integer representation from
3924	the enum AVSampleFormat in @file{libavutil/samplefmt.h}
3925
3926	@item channel_layout
3927	The channel layout of the incoming audio buffers.
3928	Either a channel layout name from channel_layout_map in
3929	@file{libavutil/channel_layout.c} or its corresponding integer representation
3930	from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
3931
3932	@item channels
3933	The number of channels of the incoming audio buffers.
3934	If both @var{channels} and @var{channel_layout} are specified, then they
3935	must be consistent.
3936
3937	@end table
3938
3939	@subsection Examples
3940
3941	@example
3942	abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
3943	@end example
3944
3945	will instruct the source to accept planar 16bit signed stereo at 44100Hz.
3946	Since the sample format with name "s16p" corresponds to the number
3947	6 and the "stereo" channel layout corresponds to the value 0x3, this is
3948	equivalent to:
3949	@example
3950	abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
3951	@end example
3952
3953	@section aevalsrc
3954
3955	Generate an audio signal specified by an expression.
3956
3957	This source accepts in input one or more expressions (one for each
3958	channel), which are evaluated and used to generate a corresponding
3959	audio signal.
3960
3961	This source accepts the following options:
3962
3963	@table @option
3964	@item exprs
3965	Set the '|'-separated expressions list for each separate channel. In case the
3966	@option{channel_layout} option is not specified, the selected channel layout
3967	depends on the number of provided expressions. Otherwise the last
3968	specified expression is applied to the remaining output channels.
3969
3970	@item channel_layout, c
3971	Set the channel layout. The number of channels in the specified layout
3972	must be equal to the number of specified expressions.
3973
3974	@item duration, d
3975	Set the minimum duration of the sourced audio. See
3976	@ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
3977	for the accepted syntax.
3978	Note that the resulting duration may be greater than the specified
3979	duration, as the generated audio is always cut at the end of a
3980	complete frame.
3981
3982	If not specified, or the expressed duration is negative, the audio is
3983	supposed to be generated forever.
3984
3985	@item nb_samples, n
3986	Set the number of samples per channel per each output frame,
3987	default to 1024.
3988
3989	@item sample_rate, s
3990	Specify the sample rate, default to 44100.
3991	@end table
3992
3993	Each expression in @var{exprs} can contain the following constants:
3994
3995	@table @option
3996	@item n
3997	number of the evaluated sample, starting from 0
3998
3999	@item t
4000	time of the evaluated sample expressed in seconds, starting from 0
4001
4002	@item s
4003	sample rate
4004
4005	@end table
4006
4007	@subsection Examples
4008
4009	@itemize
4010	@item
4011	Generate silence:
4012	@example
4013	aevalsrc=0
4014	@end example
4015
4016	@item
4017	Generate a sin signal with frequency of 440 Hz, set sample rate to
4018	8000 Hz:
4019	@example
4020	aevalsrc="sin(440*2*PI*t):s=8000"
4021	@end example
4022
4023	@item
4024	Generate a two channels signal, specify the channel layout (Front
4025	Center + Back Center) explicitly:
4026	@example
4027	aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
4028	@end example
4029
4030	@item
4031	Generate white noise:
4032	@example
4033	aevalsrc="-2+random(0)"
4034	@end example
4035
4036	@item
4037	Generate an amplitude modulated signal:
4038	@example
4039	aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
4040	@end example
4041
4042	@item
4043	Generate 2.5 Hz binaural beats on a 360 Hz carrier:
4044	@example
4045	aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
4046	@end example
4047
4048	@end itemize
4049
4050	@section anullsrc
4051
4052	The null audio source, return unprocessed audio frames. It is mainly useful
4053	as a template and to be employed in analysis / debugging tools, or as
4054	the source for filters which ignore the input data (for example the sox
4055	synth filter).
4056
4057	This source accepts the following options:
4058
4059	@table @option
4060
4061	@item channel_layout, cl
4062
4063	Specifies the channel layout, and can be either an integer or a string
4064	representing a channel layout. The default value of @var{channel_layout}
4065	is "stereo".
4066
4067	Check the channel_layout_map definition in
4068	@file{libavutil/channel_layout.c} for the mapping between strings and
4069	channel layout values.
4070
4071	@item sample_rate, r
4072	Specifies the sample rate, and defaults to 44100.
4073
4074	@item nb_samples, n
4075	Set the number of samples per requested frames.
4076
4077	@end table
4078
4079	@subsection Examples
4080
4081	@itemize
4082	@item
4083	Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
4084	@example
4085	anullsrc=r=48000:cl=4
4086	@end example
4087
4088	@item
4089	Do the same operation with a more obvious syntax:
4090	@example
4091	anullsrc=r=48000:cl=mono
4092	@end example
4093	@end itemize
4094
4095	All the parameters need to be explicitly defined.
4096
4097	@section flite
4098
4099	Synthesize a voice utterance using the libflite library.
4100
4101	To enable compilation of this filter you need to configure FFmpeg with
4102	@code{--enable-libflite}.
4103
4104	Note that the flite library is not thread-safe.
4105
4106	The filter accepts the following options:
4107
4108	@table @option
4109
4110	@item list_voices
4111	If set to 1, list the names of the available voices and exit
4112	immediately. Default value is 0.
4113
4114	@item nb_samples, n
4115	Set the maximum number of samples per frame. Default value is 512.
4116
4117	@item textfile
4118	Set the filename containing the text to speak.
4119
4120	@item text
4121	Set the text to speak.
4122
4123	@item voice, v
4124	Set the voice to use for the speech synthesis. Default value is
4125	@code{kal}. See also the @var{list_voices} option.
4126	@end table
4127
4128	@subsection Examples
4129
4130	@itemize
4131	@item
4132	Read from file @file{speech.txt}, and synthesize the text using the
4133	standard flite voice:
4134	@example
4135	flite=textfile=speech.txt
4136	@end example
4137
4138	@item
4139	Read the specified text selecting the @code{slt} voice:
4140	@example
4141	flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4142	@end example
4143
4144	@item
4145	Input text to ffmpeg:
4146	@example
4147	ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4148	@end example
4149
4150	@item
4151	Make @file{ffplay} speak the specified text, using @code{flite} and
4152	the @code{lavfi} device:
4153	@example
4154	ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
4155	@end example
4156	@end itemize
4157
4158	For more information about libflite, check:
4159	@url{http://www.speech.cs.cmu.edu/flite/}
4160
4161	@section anoisesrc
4162
4163	Generate a noise audio signal.
4164
4165	The filter accepts the following options:
4166
4167	@table @option
4168	@item sample_rate, r
4169	Specify the sample rate. Default value is 48000 Hz.
4170
4171	@item amplitude, a
4172	Specify the amplitude (0.0 - 1.0) of the generated audio stream. Default value
4173	is 1.0.
4174
4175	@item duration, d
4176	Specify the duration of the generated audio stream. Not specifying this option
4177	results in noise with an infinite length.
4178
4179	@item color, colour, c
4180	Specify the color of noise. Available noise colors are white, pink, and brown.
4181	Default color is white.
4182
4183	@item seed, s
4184	Specify a value used to seed the PRNG.
4185
4186	@item nb_samples, n
4187	Set the number of samples per each output frame, default is 1024.
4188	@end table
4189
4190	@subsection Examples
4191
4192	@itemize
4193
4194	@item
4195	Generate 60 seconds of pink noise, with a 44.1 kHz sampling rate and an amplitude of 0.5:
4196	@example
4197	anoisesrc=d=60:c=pink:r=44100:a=0.5
4198	@end example
4199	@end itemize
4200
4201	@section sine
4202
4203	Generate an audio signal made of a sine wave with amplitude 1/8.
4204
4205	The audio signal is bit-exact.
4206
4207	The filter accepts the following options:
4208
4209	@table @option
4210
4211	@item frequency, f
4212	Set the carrier frequency. Default is 440 Hz.
4213
4214	@item beep_factor, b
4215	Enable a periodic beep every second with frequency @var{beep_factor} times
4216	the carrier frequency. Default is 0, meaning the beep is disabled.
4217
4218	@item sample_rate, r
4219	Specify the sample rate, default is 44100.
4220
4221	@item duration, d
4222	Specify the duration of the generated audio stream.
4223
4224	@item samples_per_frame
4225	Set the number of samples per output frame.
4226
4227	The expression can contain the following constants:
4228
4229	@table @option
4230	@item n
4231	The (sequential) number of the output audio frame, starting from 0.
4232
4233	@item pts
4234	The PTS (Presentation TimeStamp) of the output audio frame,
4235	expressed in @var{TB} units.
4236
4237	@item t
4238	The PTS of the output audio frame, expressed in seconds.
4239
4240	@item TB
4241	The timebase of the output audio frames.
4242	@end table
4243
4244	Default is @code{1024}.
4245	@end table
4246
4247	@subsection Examples
4248
4249	@itemize
4250
4251	@item
4252	Generate a simple 440 Hz sine wave:
4253	@example
4254	sine
4255	@end example
4256
4257	@item
4258	Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
4259	@example
4260	sine=220:4:d=5
4261	sine=f=220:b=4:d=5
4262	sine=frequency=220:beep_factor=4:duration=5
4263	@end example
4264
4265	@item
4266	Generate a 1 kHz sine wave following @code{1602,1601,1602,1601,1602} NTSC
4267	pattern:
4268	@example
4269	sine=1000:samples_per_frame='st(0,mod(n,5)); 1602-not(not(eq(ld(0),1)+eq(ld(0),3)))'
4270	@end example
4271	@end itemize
4272
4273	@c man end AUDIO SOURCES
4274
4275	@chapter Audio Sinks
4276	@c man begin AUDIO SINKS
4277
4278	Below is a description of the currently available audio sinks.
4279
4280	@section abuffersink
4281
4282	Buffer audio frames, and make them available to the end of filter chain.
4283
4284	This sink is mainly intended for programmatic use, in particular
4285	through the interface defined in @file{libavfilter/buffersink.h}
4286	or the options system.
4287
4288	It accepts a pointer to an AVABufferSinkContext structure, which
4289	defines the incoming buffers' formats, to be passed as the opaque
4290	parameter to @code{avfilter_init_filter} for initialization.
4291	@section anullsink
4292
4293	Null audio sink; do absolutely nothing with the input audio. It is
4294	mainly useful as a template and for use in analysis / debugging
4295	tools.
4296
4297	@c man end AUDIO SINKS
4298
4299	@chapter Video Filters
4300	@c man begin VIDEO FILTERS
4301
4302	When you configure your FFmpeg build, you can disable any of the
4303	existing filters using @code{--disable-filters}.
4304	The configure output will show the video filters included in your
4305	build.
4306
4307	Below is a description of the currently available video filters.
4308
4309	@section alphaextract
4310
4311	Extract the alpha component from the input as a grayscale video. This
4312	is especially useful with the @var{alphamerge} filter.
4313
4314	@section alphamerge
4315
4316	Add or replace the alpha component of the primary input with the
4317	grayscale value of a second input. This is intended for use with
4318	@var{alphaextract} to allow the transmission or storage of frame
4319	sequences that have alpha in a format that doesn't support an alpha
4320	channel.
4321
4322	For example, to reconstruct full frames from a normal YUV-encoded video
4323	and a separate video created with @var{alphaextract}, you might use:
4324	@example
4325	movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
4326	@end example
4327
4328	Since this filter is designed for reconstruction, it operates on frame
4329	sequences without considering timestamps, and terminates when either
4330	input reaches end of stream. This will cause problems if your encoding
4331	pipeline drops frames. If you're trying to apply an image as an
4332	overlay to a video stream, consider the @var{overlay} filter instead.
4333
4334	@section ass
4335
4336	Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
4337	and libavformat to work. On the other hand, it is limited to ASS (Advanced
4338	Substation Alpha) subtitles files.
4339
4340	This filter accepts the following option in addition to the common options from
4341	the @ref{subtitles} filter:
4342
4343	@table @option
4344	@item shaping
4345	Set the shaping engine
4346
4347	Available values are:
4348	@table @samp
4349	@item auto
4350	The default libass shaping engine, which is the best available.
4351	@item simple
4352	Fast, font-agnostic shaper that can do only substitutions
4353	@item complex
4354	Slower shaper using OpenType for substitutions and positioning
4355	@end table
4356
4357	The default is @code{auto}.
4358	@end table
4359
4360	@section atadenoise
4361	Apply an Adaptive Temporal Averaging Denoiser to the video input.
4362
4363	The filter accepts the following options:
4364
4365	@table @option
4366	@item 0a
4367	Set threshold A for 1st plane. Default is 0.02.
4368	Valid range is 0 to 0.3.
4369
4370	@item 0b
4371	Set threshold B for 1st plane. Default is 0.04.
4372	Valid range is 0 to 5.
4373
4374	@item 1a
4375	Set threshold A for 2nd plane. Default is 0.02.
4376	Valid range is 0 to 0.3.
4377
4378	@item 1b
4379	Set threshold B for 2nd plane. Default is 0.04.
4380	Valid range is 0 to 5.
4381
4382	@item 2a
4383	Set threshold A for 3rd plane. Default is 0.02.
4384	Valid range is 0 to 0.3.
4385
4386	@item 2b
4387	Set threshold B for 3rd plane. Default is 0.04.
4388	Valid range is 0 to 5.
4389
4390	Threshold A is designed to react on abrupt changes in the input signal and
4391	threshold B is designed to react on continuous changes in the input signal.
4392
4393	@item s
4394	Set number of frames filter will use for averaging. Default is 33. Must be odd
4395	number in range [5, 129].
4396
4397	@item p
4398	Set what planes of frame filter will use for averaging. Default is all.
4399	@end table
4400
4401	@section avgblur
4402
4403	Apply average blur filter.
4404
4405	The filter accepts the following options:
4406
4407	@table @option
4408	@item sizeX
4409	Set horizontal kernel size.
4410
4411	@item planes
4412	Set which planes to filter. By default all planes are filtered.
4413
4414	@item sizeY
4415	Set vertical kernel size, if zero it will be same as @code{sizeX}.
4416	Default is @code{0}.
4417	@end table
4418
4419	@section bbox
4420
4421	Compute the bounding box for the non-black pixels in the input frame
4422	luminance plane.
4423
4424	This filter computes the bounding box containing all the pixels with a
4425	luminance value greater than the minimum allowed value.
4426	The parameters describing the bounding box are printed on the filter
4427	log.
4428
4429	The filter accepts the following option:
4430
4431	@table @option
4432	@item min_val
4433	Set the minimal luminance value. Default is @code{16}.
4434	@end table
4435
4436	@section bitplanenoise
4437
4438	Show and measure bit plane noise.
4439
4440	The filter accepts the following options:
4441
4442	@table @option
4443	@item bitplane
4444	Set which plane to analyze. Default is @code{1}.
4445
4446	@item filter
4447	Filter out noisy pixels from @code{bitplane} set above.
4448	Default is disabled.
4449	@end table
4450
4451	@section blackdetect
4452
4453	Detect video intervals that are (almost) completely black. Can be
4454	useful to detect chapter transitions, commercials, or invalid
4455	recordings. Output lines contains the time for the start, end and
4456	duration of the detected black interval expressed in seconds.
4457
4458	In order to display the output lines, you need to set the loglevel at
4459	least to the AV_LOG_INFO value.
4460
4461	The filter accepts the following options:
4462
4463	@table @option
4464	@item black_min_duration, d
4465	Set the minimum detected black duration expressed in seconds. It must
4466	be a non-negative floating point number.
4467
4468	Default value is 2.0.
4469
4470	@item picture_black_ratio_th, pic_th
4471	Set the threshold for considering a picture "black".
4472	Express the minimum value for the ratio:
4473	@example
4474	@var{nb_black_pixels} / @var{nb_pixels}
4475	@end example
4476
4477	for which a picture is considered black.
4478	Default value is 0.98.
4479
4480	@item pixel_black_th, pix_th
4481	Set the threshold for considering a pixel "black".
4482
4483	The threshold expresses the maximum pixel luminance value for which a
4484	pixel is considered "black". The provided value is scaled according to
4485	the following equation:
4486	@example
4487	@var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
4488	@end example
4489
4490	@var{luminance_range_size} and @var{luminance_minimum_value} depend on
4491	the input video format, the range is [0-255] for YUV full-range
4492	formats and [16-235] for YUV non full-range formats.
4493
4494	Default value is 0.10.
4495	@end table
4496
4497	The following example sets the maximum pixel threshold to the minimum
4498	value, and detects only black intervals of 2 or more seconds:
4499	@example
4500	blackdetect=d=2:pix_th=0.00
4501	@end example
4502
4503	@section blackframe
4504
4505	Detect frames that are (almost) completely black. Can be useful to
4506	detect chapter transitions or commercials. Output lines consist of
4507	the frame number of the detected frame, the percentage of blackness,
4508	the position in the file if known or -1 and the timestamp in seconds.
4509
4510	In order to display the output lines, you need to set the loglevel at
4511	least to the AV_LOG_INFO value.
4512
4513	This filter exports frame metadata @code{lavfi.blackframe.pblack}.
4514	The value represents the percentage of pixels in the picture that
4515	are below the threshold value.
4516
4517	It accepts the following parameters:
4518
4519	@table @option
4520
4521	@item amount
4522	The percentage of the pixels that have to be below the threshold; it defaults to
4523	@code{98}.
4524
4525	@item threshold, thresh
4526	The threshold below which a pixel value is considered black; it defaults to
4527	@code{32}.
4528
4529	@end table
4530
4531	@section blend, tblend
4532
4533	Blend two video frames into each other.
4534
4535	The @code{blend} filter takes two input streams and outputs one
4536	stream, the first input is the "top" layer and second input is
4537	"bottom" layer. By default, the output terminates when the longest input terminates.
4538
4539	The @code{tblend} (time blend) filter takes two consecutive frames
4540	from one single stream, and outputs the result obtained by blending
4541	the new frame on top of the old frame.
4542
4543	A description of the accepted options follows.
4544
4545	@table @option
4546	@item c0_mode
4547	@item c1_mode
4548	@item c2_mode
4549	@item c3_mode
4550	@item all_mode
4551	Set blend mode for specific pixel component or all pixel components in case
4552	of @var{all_mode}. Default value is @code{normal}.
4553
4554	Available values for component modes are:
4555	@table @samp
4556	@item addition
4557	@item addition128
4558	@item and
4559	@item average
4560	@item burn
4561	@item darken
4562	@item difference
4563	@item difference128
4564	@item divide
4565	@item dodge
4566	@item freeze
4567	@item exclusion
4568	@item glow
4569	@item hardlight
4570	@item hardmix
4571	@item heat
4572	@item lighten
4573	@item linearlight
4574	@item multiply
4575	@item multiply128
4576	@item negation
4577	@item normal
4578	@item or
4579	@item overlay
4580	@item phoenix
4581	@item pinlight
4582	@item reflect
4583	@item screen
4584	@item softlight
4585	@item subtract
4586	@item vividlight
4587	@item xor
4588	@end table
4589
4590	@item c0_opacity
4591	@item c1_opacity
4592	@item c2_opacity
4593	@item c3_opacity
4594	@item all_opacity
4595	Set blend opacity for specific pixel component or all pixel components in case
4596	of @var{all_opacity}. Only used in combination with pixel component blend modes.
4597
4598	@item c0_expr
4599	@item c1_expr
4600	@item c2_expr
4601	@item c3_expr
4602	@item all_expr
4603	Set blend expression for specific pixel component or all pixel components in case
4604	of @var{all_expr}. Note that related mode options will be ignored if those are set.
4605
4606	The expressions can use the following variables:
4607
4608	@table @option
4609	@item N
4610	The sequential number of the filtered frame, starting from @code{0}.
4611
4612	@item X
4613	@item Y
4614	the coordinates of the current sample
4615
4616	@item W
4617	@item H
4618	the width and height of currently filtered plane
4619
4620	@item SW
4621	@item SH
4622	Width and height scale depending on the currently filtered plane. It is the
4623	ratio between the corresponding luma plane number of pixels and the current
4624	plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
4625	@code{0.5,0.5} for chroma planes.
4626
4627	@item T
4628	Time of the current frame, expressed in seconds.
4629
4630	@item TOP, A
4631	Value of pixel component at current location for first video frame (top layer).
4632
4633	@item BOTTOM, B
4634	Value of pixel component at current location for second video frame (bottom layer).
4635	@end table
4636
4637	@item shortest
4638	Force termination when the shortest input terminates. Default is
4639	@code{0}. This option is only defined for the @code{blend} filter.
4640
4641	@item repeatlast
4642	Continue applying the last bottom frame after the end of the stream. A value of
4643	@code{0} disable the filter after the last frame of the bottom layer is reached.
4644	Default is @code{1}. This option is only defined for the @code{blend} filter.
4645	@end table
4646
4647	@subsection Examples
4648
4649	@itemize
4650	@item
4651	Apply transition from bottom layer to top layer in first 10 seconds:
4652	@example
4653	blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
4654	@end example
4655
4656	@item
4657	Apply 1x1 checkerboard effect:
4658	@example
4659	blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
4660	@end example
4661
4662	@item
4663	Apply uncover left effect:
4664	@example
4665	blend=all_expr='if(gte(N*SW+X,W),A,B)'
4666	@end example
4667
4668	@item
4669	Apply uncover down effect:
4670	@example
4671	blend=all_expr='if(gte(Y-N*SH,0),A,B)'
4672	@end example
4673
4674	@item
4675	Apply uncover up-left effect:
4676	@example
4677	blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
4678	@end example
4679
4680	@item
4681	Split diagonally video and shows top and bottom layer on each side:
4682	@example
4683	blend=all_expr=if(gt(X,Y*(W/H)),A,B)
4684	@end example
4685
4686	@item
4687	Display differences between the current and the previous frame:
4688	@example
4689	tblend=all_mode=difference128
4690	@end example
4691	@end itemize
4692
4693	@section boxblur
4694
4695	Apply a boxblur algorithm to the input video.
4696
4697	It accepts the following parameters:
4698
4699	@table @option
4700
4701	@item luma_radius, lr
4702	@item luma_power, lp
4703	@item chroma_radius, cr
4704	@item chroma_power, cp
4705	@item alpha_radius, ar
4706	@item alpha_power, ap
4707
4708	@end table
4709
4710	A description of the accepted options follows.
4711
4712	@table @option
4713	@item luma_radius, lr
4714	@item chroma_radius, cr
4715	@item alpha_radius, ar
4716	Set an expression for the box radius in pixels used for blurring the
4717	corresponding input plane.
4718
4719	The radius value must be a non-negative number, and must not be
4720	greater than the value of the expression @code{min(w,h)/2} for the
4721	luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
4722	planes.
4723
4724	Default value for @option{luma_radius} is "2". If not specified,
4725	@option{chroma_radius} and @option{alpha_radius} default to the
4726	corresponding value set for @option{luma_radius}.
4727
4728	The expressions can contain the following constants:
4729	@table @option
4730	@item w
4731	@item h
4732	The input width and height in pixels.
4733
4734	@item cw
4735	@item ch
4736	The input chroma image width and height in pixels.
4737
4738	@item hsub
4739	@item vsub
4740	The horizontal and vertical chroma subsample values. For example, for the
4741	pixel format "yuv422p", @var{hsub} is 2 and @var{vsub} is 1.
4742	@end table
4743
4744	@item luma_power, lp
4745	@item chroma_power, cp
4746	@item alpha_power, ap
4747	Specify how many times the boxblur filter is applied to the
4748	corresponding plane.
4749
4750	Default value for @option{luma_power} is 2. If not specified,
4751	@option{chroma_power} and @option{alpha_power} default to the
4752	corresponding value set for @option{luma_power}.
4753
4754	A value of 0 will disable the effect.
4755	@end table
4756
4757	@subsection Examples
4758
4759	@itemize
4760	@item
4761	Apply a boxblur filter with the luma, chroma, and alpha radii
4762	set to 2:
4763	@example
4764	boxblur=luma_radius=2:luma_power=1
4765	boxblur=2:1
4766	@end example
4767
4768	@item
4769	Set the luma radius to 2, and alpha and chroma radius to 0:
4770	@example
4771	boxblur=2:1:cr=0:ar=0
4772	@end example
4773
4774	@item
4775	Set the luma and chroma radii to a fraction of the video dimension:
4776	@example
4777	boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
4778	@end example
4779	@end itemize
4780
4781	@section bwdif
4782
4783	Deinterlace the input video ("bwdif" stands for "Bob Weaver
4784	Deinterlacing Filter").
4785
4786	Motion adaptive deinterlacing based on yadif with the use of w3fdif and cubic
4787	interpolation algorithms.
4788	It accepts the following parameters:
4789
4790	@table @option
4791	@item mode
4792	The interlacing mode to adopt. It accepts one of the following values:
4793
4794	@table @option
4795	@item 0, send_frame
4796	Output one frame for each frame.
4797	@item 1, send_field
4798	Output one frame for each field.
4799	@end table
4800
4801	The default value is @code{send_field}.
4802
4803	@item parity
4804	The picture field parity assumed for the input interlaced video. It accepts one
4805	of the following values:
4806
4807	@table @option
4808	@item 0, tff
4809	Assume the top field is first.
4810	@item 1, bff
4811	Assume the bottom field is first.
4812	@item -1, auto
4813	Enable automatic detection of field parity.
4814	@end table
4815
4816	The default value is @code{auto}.
4817	If the interlacing is unknown or the decoder does not export this information,
4818	top field first will be assumed.
4819
4820	@item deint
4821	Specify which frames to deinterlace. Accept one of the following
4822	values:
4823
4824	@table @option
4825	@item 0, all
4826	Deinterlace all frames.
4827	@item 1, interlaced
4828	Only deinterlace frames marked as interlaced.
4829	@end table
4830
4831	The default value is @code{all}.
4832	@end table
4833
4834	@section chromakey
4835	YUV colorspace color/chroma keying.
4836
4837	The filter accepts the following options:
4838
4839	@table @option
4840	@item color
4841	The color which will be replaced with transparency.
4842
4843	@item similarity
4844	Similarity percentage with the key color.
4845
4846	0.01 matches only the exact key color, while 1.0 matches everything.
4847
4848	@item blend
4849	Blend percentage.
4850
4851	0.0 makes pixels either fully transparent, or not transparent at all.
4852
4853	Higher values result in semi-transparent pixels, with a higher transparency
4854	the more similar the pixels color is to the key color.
4855
4856	@item yuv
4857	Signals that the color passed is already in YUV instead of RGB.
4858
4859	Litteral colors like "green" or "red" don't make sense with this enabled anymore.
4860	This can be used to pass exact YUV values as hexadecimal numbers.
4861	@end table
4862
4863	@subsection Examples
4864
4865	@itemize
4866	@item
4867	Make every green pixel in the input image transparent:
4868	@example
4869	ffmpeg -i input.png -vf chromakey=green out.png
4870	@end example
4871
4872	@item
4873	Overlay a greenscreen-video on top of a static black background.
4874	@example
4875	ffmpeg -f lavfi -i color=c=black:s=1280x720 -i video.mp4 -shortest -filter_complex "[1:v]chromakey=0x70de77:0.1:0.2[ckout];[0:v][ckout]overlay[out]" -map "[out]" output.mkv
4876	@end example
4877	@end itemize
4878
4879	@section ciescope
4880
4881	Display CIE color diagram with pixels overlaid onto it.
4882
4883	The filter accepts the following options:
4884
4885	@table @option
4886	@item system
4887	Set color system.
4888
4889	@table @samp
4890	@item ntsc, 470m
4891	@item ebu, 470bg
4892	@item smpte
4893	@item 240m
4894	@item apple
4895	@item widergb
4896	@item cie1931
4897	@item rec709, hdtv
4898	@item uhdtv, rec2020
4899	@end table
4900
4901	@item cie
4902	Set CIE system.
4903
4904	@table @samp
4905	@item xyy
4906	@item ucs
4907	@item luv
4908	@end table
4909
4910	@item gamuts
4911	Set what gamuts to draw.
4912
4913	See @code{system} option for available values.
4914
4915	@item size, s
4916	Set ciescope size, by default set to 512.
4917
4918	@item intensity, i
4919	Set intensity used to map input pixel values to CIE diagram.
4920
4921	@item contrast
4922	Set contrast used to draw tongue colors that are out of active color system gamut.
4923
4924	@item corrgamma
4925	Correct gamma displayed on scope, by default enabled.
4926
4927	@item showwhite
4928	Show white point on CIE diagram, by default disabled.
4929
4930	@item gamma
4931	Set input gamma. Used only with XYZ input color space.
4932	@end table
4933
4934	@section codecview
4935
4936	Visualize information exported by some codecs.
4937
4938	Some codecs can export information through frames using side-data or other
4939	means. For example, some MPEG based codecs export motion vectors through the
4940	@var{export_mvs} flag in the codec @option{flags2} option.
4941
4942	The filter accepts the following option:
4943
4944	@table @option
4945	@item mv
4946	Set motion vectors to visualize.
4947
4948	Available flags for @var{mv} are:
4949
4950	@table @samp
4951	@item pf
4952	forward predicted MVs of P-frames
4953	@item bf
4954	forward predicted MVs of B-frames
4955	@item bb
4956	backward predicted MVs of B-frames
4957	@end table
4958
4959	@item qp
4960	Display quantization parameters using the chroma planes.
4961
4962	@item mv_type, mvt
4963	Set motion vectors type to visualize. Includes MVs from all frames unless specified by @var{frame_type} option.
4964
4965	Available flags for @var{mv_type} are:
4966
4967	@table @samp
4968	@item fp
4969	forward predicted MVs
4970	@item bp
4971	backward predicted MVs
4972	@end table
4973
4974	@item frame_type, ft
4975	Set frame type to visualize motion vectors of.
4976
4977	Available flags for @var{frame_type} are:
4978
4979	@table @samp
4980	@item if
4981	intra-coded frames (I-frames)
4982	@item pf
4983	predicted frames (P-frames)
4984	@item bf
4985	bi-directionally predicted frames (B-frames)
4986	@end table
4987	@end table
4988
4989	@subsection Examples
4990
4991	@itemize
4992	@item
4993	Visualize forward predicted MVs of all frames using @command{ffplay}:
4994	@example
4995	ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv_type=fp
4996	@end example
4997
4998	@item
4999	Visualize multi-directionals MVs of P and B-Frames using @command{ffplay}:
5000	@example
5001	ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv=pf+bf+bb
5002	@end example
5003	@end itemize
5004
5005	@section colorbalance
5006	Modify intensity of primary colors (red, green and blue) of input frames.
5007
5008	The filter allows an input frame to be adjusted in the shadows, midtones or highlights
5009	regions for the red-cyan, green-magenta or blue-yellow balance.
5010
5011	A positive adjustment value shifts the balance towards the primary color, a negative
5012	value towards the complementary color.
5013
5014	The filter accepts the following options:
5015
5016	@table @option
5017	@item rs
5018	@item gs
5019	@item bs
5020	Adjust red, green and blue shadows (darkest pixels).
5021
5022	@item rm
5023	@item gm
5024	@item bm
5025	Adjust red, green and blue midtones (medium pixels).
5026
5027	@item rh
5028	@item gh
5029	@item bh
5030	Adjust red, green and blue highlights (brightest pixels).
5031
5032	Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
5033	@end table
5034
5035	@subsection Examples
5036
5037	@itemize
5038	@item
5039	Add red color cast to shadows:
5040	@example
5041	colorbalance=rs=.3
5042	@end example
5043	@end itemize
5044
5045	@section colorkey
5046	RGB colorspace color keying.
5047
5048	The filter accepts the following options:
5049
5050	@table @option
5051	@item color
5052	The color which will be replaced with transparency.
5053
5054	@item similarity
5055	Similarity percentage with the key color.
5056
5057	0.01 matches only the exact key color, while 1.0 matches everything.
5058
5059	@item blend
5060	Blend percentage.
5061
5062	0.0 makes pixels either fully transparent, or not transparent at all.
5063
5064	Higher values result in semi-transparent pixels, with a higher transparency
5065	the more similar the pixels color is to the key color.
5066	@end table
5067
5068	@subsection Examples
5069
5070	@itemize
5071	@item
5072	Make every green pixel in the input image transparent:
5073	@example
5074	ffmpeg -i input.png -vf colorkey=green out.png
5075	@end example
5076
5077	@item
5078	Overlay a greenscreen-video on top of a static background image.
5079	@example
5080	ffmpeg -i background.png -i video.mp4 -filter_complex "[1:v]colorkey=0x3BBD1E:0.3:0.2[ckout];[0:v][ckout]overlay[out]" -map "[out]" output.flv
5081	@end example
5082	@end itemize
5083
5084	@section colorlevels
5085
5086	Adjust video input frames using levels.
5087
5088	The filter accepts the following options:
5089
5090	@table @option
5091	@item rimin
5092	@item gimin
5093	@item bimin
5094	@item aimin
5095	Adjust red, green, blue and alpha input black point.
5096	Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
5097
5098	@item rimax
5099	@item gimax
5100	@item bimax
5101	@item aimax
5102	Adjust red, green, blue and alpha input white point.
5103	Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{1}.
5104
5105	Input levels are used to lighten highlights (bright tones), darken shadows
5106	(dark tones), change the balance of bright and dark tones.
5107
5108	@item romin
5109	@item gomin
5110	@item bomin
5111	@item aomin
5112	Adjust red, green, blue and alpha output black point.
5113	Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{0}.
5114
5115	@item romax
5116	@item gomax
5117	@item bomax
5118	@item aomax
5119	Adjust red, green, blue and alpha output white point.
5120	Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{1}.
5121
5122	Output levels allows manual selection of a constrained output level range.
5123	@end table
5124
5125	@subsection Examples
5126
5127	@itemize
5128	@item
5129	Make video output darker:
5130	@example
5131	colorlevels=rimin=0.058:gimin=0.058:bimin=0.058
5132	@end example
5133
5134	@item
5135	Increase contrast:
5136	@example
5137	colorlevels=rimin=0.039:gimin=0.039:bimin=0.039:rimax=0.96:gimax=0.96:bimax=0.96
5138	@end example
5139
5140	@item
5141	Make video output lighter:
5142	@example
5143	colorlevels=rimax=0.902:gimax=0.902:bimax=0.902
5144	@end example
5145
5146	@item
5147	Increase brightness:
5148	@example
5149	colorlevels=romin=0.5:gomin=0.5:bomin=0.5
5150	@end example
5151	@end itemize
5152
5153	@section colorchannelmixer
5154
5155	Adjust video input frames by re-mixing color channels.
5156
5157	This filter modifies a color channel by adding the values associated to
5158	the other channels of the same pixels. For example if the value to
5159	modify is red, the output value will be:
5160	@example
5161	@var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
5162	@end example
5163
5164	The filter accepts the following options:
5165
5166	@table @option
5167	@item rr
5168	@item rg
5169	@item rb
5170	@item ra
5171	Adjust contribution of input red, green, blue and alpha channels for output red channel.
5172	Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
5173
5174	@item gr
5175	@item gg
5176	@item gb
5177	@item ga
5178	Adjust contribution of input red, green, blue and alpha channels for output green channel.
5179	Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
5180
5181	@item br
5182	@item bg
5183	@item bb
5184	@item ba
5185	Adjust contribution of input red, green, blue and alpha channels for output blue channel.
5186	Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
5187
5188	@item ar
5189	@item ag
5190	@item ab
5191	@item aa
5192	Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
5193	Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
5194
5195	Allowed ranges for options are @code{[-2.0, 2.0]}.
5196	@end table
5197
5198	@subsection Examples
5199
5200	@itemize
5201	@item
5202	Convert source to grayscale:
5203	@example
5204	colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
5205	@end example
5206	@item
5207	Simulate sepia tones:
5208	@example
5209	colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
5210	@end example
5211	@end itemize
5212
5213	@section colormatrix
5214
5215	Convert color matrix.
5216
5217	The filter accepts the following options:
5218
5219	@table @option
5220	@item src
5221	@item dst
5222	Specify the source and destination color matrix. Both values must be
5223	specified.
5224
5225	The accepted values are:
5226	@table @samp
5227	@item bt709
5228	BT.709
5229
5230	@item fcc
5231	FCC
5232
5233	@item bt601
5234	BT.601
5235
5236	@item bt470
5237	BT.470
5238
5239	@item bt470bg
5240	BT.470BG
5241
5242	@item smpte170m
5243	SMPTE-170M
5244
5245	@item smpte240m
5246	SMPTE-240M
5247
5248	@item bt2020
5249	BT.2020
5250	@end table
5251	@end table
5252
5253	For example to convert from BT.601 to SMPTE-240M, use the command:
5254	@example
5255	colormatrix=bt601:smpte240m
5256	@end example
5257
5258	@section colorspace
5259
5260	Convert colorspace, transfer characteristics or color primaries.
5261	Input video needs to have an even size.
5262
5263	The filter accepts the following options:
5264
5265	@table @option
5266	@anchor{all}
5267	@item all
5268	Specify all color properties at once.
5269
5270	The accepted values are:
5271	@table @samp
5272	@item bt470m
5273	BT.470M
5274
5275	@item bt470bg
5276	BT.470BG
5277
5278	@item bt601-6-525
5279	BT.601-6 525
5280
5281	@item bt601-6-625
5282	BT.601-6 625
5283
5284	@item bt709
5285	BT.709
5286
5287	@item smpte170m
5288	SMPTE-170M
5289
5290	@item smpte240m
5291	SMPTE-240M
5292
5293	@item bt2020
5294	BT.2020
5295
5296	@end table
5297
5298	@anchor{space}
5299	@item space
5300	Specify output colorspace.
5301
5302	The accepted values are:
5303	@table @samp
5304	@item bt709
5305	BT.709
5306
5307	@item fcc
5308	FCC
5309
5310	@item bt470bg
5311	BT.470BG or BT.601-6 625
5312
5313	@item smpte170m
5314	SMPTE-170M or BT.601-6 525
5315
5316	@item smpte240m
5317	SMPTE-240M
5318
5319	@item ycgco
5320	YCgCo
5321
5322	@item bt2020ncl
5323	BT.2020 with non-constant luminance
5324
5325	@end table
5326
5327	@anchor{trc}
5328	@item trc
5329	Specify output transfer characteristics.
5330
5331	The accepted values are:
5332	@table @samp
5333	@item bt709
5334	BT.709
5335
5336	@item bt470m
5337	BT.470M
5338
5339	@item bt470bg
5340	BT.470BG
5341
5342	@item gamma22
5343	Constant gamma of 2.2
5344
5345	@item gamma28
5346	Constant gamma of 2.8
5347
5348	@item smpte170m
5349	SMPTE-170M, BT.601-6 625 or BT.601-6 525
5350
5351	@item smpte240m
5352	SMPTE-240M
5353
5354	@item srgb
5355	SRGB
5356
5357	@item iec61966-2-1
5358	iec61966-2-1
5359
5360	@item iec61966-2-4
5361	iec61966-2-4
5362
5363	@item xvycc
5364	xvycc
5365
5366	@item bt2020-10
5367	BT.2020 for 10-bits content
5368
5369	@item bt2020-12
5370	BT.2020 for 12-bits content
5371
5372	@end table
5373
5374	@anchor{primaries}
5375	@item primaries
5376	Specify output color primaries.
5377
5378	The accepted values are:
5379	@table @samp
5380	@item bt709
5381	BT.709
5382
5383	@item bt470m
5384	BT.470M
5385
5386	@item bt470bg
5387	BT.470BG or BT.601-6 625
5388
5389	@item smpte170m
5390	SMPTE-170M or BT.601-6 525
5391
5392	@item smpte240m
5393	SMPTE-240M
5394
5395	@item film
5396	film
5397
5398	@item smpte431
5399	SMPTE-431
5400
5401	@item smpte432
5402	SMPTE-432
5403
5404	@item bt2020
5405	BT.2020
5406
5407	@end table
5408
5409	@anchor{range}
5410	@item range
5411	Specify output color range.
5412
5413	The accepted values are:
5414	@table @samp
5415	@item tv
5416	TV (restricted) range
5417
5418	@item mpeg
5419	MPEG (restricted) range
5420
5421	@item pc
5422	PC (full) range
5423
5424	@item jpeg
5425	JPEG (full) range
5426
5427	@end table
5428
5429	@item format
5430	Specify output color format.
5431
5432	The accepted values are:
5433	@table @samp
5434	@item yuv420p
5435	YUV 4:2:0 planar 8-bits
5436
5437	@item yuv420p10
5438	YUV 4:2:0 planar 10-bits
5439
5440	@item yuv420p12
5441	YUV 4:2:0 planar 12-bits
5442
5443	@item yuv422p
5444	YUV 4:2:2 planar 8-bits
5445
5446	@item yuv422p10
5447	YUV 4:2:2 planar 10-bits
5448
5449	@item yuv422p12
5450	YUV 4:2:2 planar 12-bits
5451
5452	@item yuv444p
5453	YUV 4:4:4 planar 8-bits
5454
5455	@item yuv444p10
5456	YUV 4:4:4 planar 10-bits
5457
5458	@item yuv444p12
5459	YUV 4:4:4 planar 12-bits
5460
5461	@end table
5462
5463	@item fast
5464	Do a fast conversion, which skips gamma/primary correction. This will take
5465	significantly less CPU, but will be mathematically incorrect. To get output
5466	compatible with that produced by the colormatrix filter, use fast=1.
5467
5468	@item dither
5469	Specify dithering mode.
5470
5471	The accepted values are:
5472	@table @samp
5473	@item none
5474	No dithering
5475
5476	@item fsb
5477	Floyd-Steinberg dithering
5478	@end table
5479
5480	@item wpadapt
5481	Whitepoint adaptation mode.
5482
5483	The accepted values are:
5484	@table @samp
5485	@item bradford
5486	Bradford whitepoint adaptation
5487
5488	@item vonkries
5489	von Kries whitepoint adaptation
5490
5491	@item identity
5492	identity whitepoint adaptation (i.e. no whitepoint adaptation)
5493	@end table
5494
5495	@item iall
5496	Override all input properties at once. Same accepted values as @ref{all}.
5497
5498	@item ispace
5499	Override input colorspace. Same accepted values as @ref{space}.
5500
5501	@item iprimaries
5502	Override input color primaries. Same accepted values as @ref{primaries}.
5503
5504	@item itrc
5505	Override input transfer characteristics. Same accepted values as @ref{trc}.
5506
5507	@item irange
5508	Override input color range. Same accepted values as @ref{range}.
5509
5510	@end table
5511
5512	The filter converts the transfer characteristics, color space and color
5513	primaries to the specified user values. The output value, if not specified,
5514	is set to a default value based on the "all" property. If that property is
5515	also not specified, the filter will log an error. The output color range and
5516	format default to the same value as the input color range and format. The
5517	input transfer characteristics, color space, color primaries and color range
5518	should be set on the input data. If any of these are missing, the filter will
5519	log an error and no conversion will take place.
5520
5521	For example to convert the input to SMPTE-240M, use the command:
5522	@example
5523	colorspace=smpte240m
5524	@end example
5525
5526	@section convolution
5527
5528	Apply convolution 3x3 or 5x5 filter.
5529
5530	The filter accepts the following options:
5531
5532	@table @option
5533	@item 0m
5534	@item 1m
5535	@item 2m
5536	@item 3m
5537	Set matrix for each plane.
5538	Matrix is sequence of 9 or 25 signed integers.
5539
5540	@item 0rdiv
5541	@item 1rdiv
5542	@item 2rdiv
5543	@item 3rdiv
5544	Set multiplier for calculated value for each plane.
5545
5546	@item 0bias
5547	@item 1bias
5548	@item 2bias
5549	@item 3bias
5550	Set bias for each plane. This value is added to the result of the multiplication.
5551	Useful for making the overall image brighter or darker. Default is 0.0.
5552	@end table
5553
5554	@subsection Examples
5555
5556	@itemize
5557	@item
5558	Apply sharpen:
5559	@example
5560	convolution="0 -1 0 -1 5 -1 0 -1 0:0 -1 0 -1 5 -1 0 -1 0:0 -1 0 -1 5 -1 0 -1 0:0 -1 0 -1 5 -1 0 -1 0"
5561	@end example
5562
5563	@item
5564	Apply blur:
5565	@example
5566	convolution="1 1 1 1 1 1 1 1 1:1 1 1 1 1 1 1 1 1:1 1 1 1 1 1 1 1 1:1 1 1 1 1 1 1 1 1:1/9:1/9:1/9:1/9"
5567	@end example
5568
5569	@item
5570	Apply edge enhance:
5571	@example
5572	convolution="0 0 0 -1 1 0 0 0 0:0 0 0 -1 1 0 0 0 0:0 0 0 -1 1 0 0 0 0:0 0 0 -1 1 0 0 0 0:5:1:1:1:0:128:128:128"
5573	@end example
5574
5575	@item
5576	Apply edge detect:
5577	@example
5578	convolution="0 1 0 1 -4 1 0 1 0:0 1 0 1 -4 1 0 1 0:0 1 0 1 -4 1 0 1 0:0 1 0 1 -4 1 0 1 0:5:5:5:1:0:128:128:128"
5579	@end example
5580
5581	@item
5582	Apply emboss:
5583	@example
5584	convolution="-2 -1 0 -1 1 1 0 1 2:-2 -1 0 -1 1 1 0 1 2:-2 -1 0 -1 1 1 0 1 2:-2 -1 0 -1 1 1 0 1 2"
5585	@end example
5586	@end itemize
5587
5588	@section copy
5589
5590	Copy the input source unchanged to the output. This is mainly useful for
5591	testing purposes.
5592
5593	@anchor{coreimage}
5594	@section coreimage
5595	Video filtering on GPU using Apple's CoreImage API on OSX.
5596
5597	Hardware acceleration is based on an OpenGL context. Usually, this means it is
5598	processed by video hardware. However, software-based OpenGL implementations
5599	exist which means there is no guarantee for hardware processing. It depends on
5600	the respective OSX.
5601
5602	There are many filters and image generators provided by Apple that come with a
5603	large variety of options. The filter has to be referenced by its name along
5604	with its options.
5605
5606	The coreimage filter accepts the following options:
5607	@table @option
5608	@item list_filters
5609	List all available filters and generators along with all their respective
5610	options as well as possible minimum and maximum values along with the default
5611	values.
5612	@example
5613	list_filters=true
5614	@end example
5615
5616	@item filter
5617	Specify all filters by their respective name and options.
5618	Use @var{list_filters} to determine all valid filter names and options.
5619	Numerical options are specified by a float value and are automatically clamped
5620	to their respective value range. Vector and color options have to be specified
5621	by a list of space separated float values. Character escaping has to be done.
5622	A special option name @code{default} is available to use default options for a
5623	filter.
5624
5625	It is required to specify either @code{default} or at least one of the filter options.
5626	All omitted options are used with their default values.
5627	The syntax of the filter string is as follows:
5628	@example
5629	filter=<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...][#<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...]][#...]
5630	@end example
5631
5632	@item output_rect
5633	Specify a rectangle where the output of the filter chain is copied into the
5634	input image. It is given by a list of space separated float values:
5635	@example
5636	output_rect=x\ y\ width\ height
5637	@end example
5638	If not given, the output rectangle equals the dimensions of the input image.
5639	The output rectangle is automatically cropped at the borders of the input
5640	image. Negative values are valid for each component.
5641	@example
5642	output_rect=25\ 25\ 100\ 100
5643	@end example
5644	@end table
5645
5646	Several filters can be chained for successive processing without GPU-HOST
5647	transfers allowing for fast processing of complex filter chains.
5648	Currently, only filters with zero (generators) or exactly one (filters) input
5649	image and one output image are supported. Also, transition filters are not yet
5650	usable as intended.
5651
5652	Some filters generate output images with additional padding depending on the
5653	respective filter kernel. The padding is automatically removed to ensure the
5654	filter output has the same size as the input image.
5655
5656	For image generators, the size of the output image is determined by the
5657	previous output image of the filter chain or the input image of the whole
5658	filterchain, respectively. The generators do not use the pixel information of
5659	this image to generate their output. However, the generated output is
5660	blended onto this image, resulting in partial or complete coverage of the
5661	output image.
5662
5663	The @ref{coreimagesrc} video source can be used for generating input images
5664	which are directly fed into the filter chain. By using it, providing input
5665	images by another video source or an input video is not required.
5666
5667	@subsection Examples
5668
5669	@itemize
5670
5671	@item
5672	List all filters available:
5673	@example
5674	coreimage=list_filters=true
5675	@end example
5676
5677	@item
5678	Use the CIBoxBlur filter with default options to blur an image:
5679	@example
5680	coreimage=filter=CIBoxBlur@@default
5681	@end example
5682
5683	@item
5684	Use a filter chain with CISepiaTone at default values and CIVignetteEffect with
5685	its center at 100x100 and a radius of 50 pixels:
5686	@example
5687	coreimage=filter=CIBoxBlur@@default#CIVignetteEffect@@inputCenter=100\ 100@@inputRadius=50
5688	@end example
5689
5690	@item
5691	Use nullsrc and CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
5692	given as complete and escaped command-line for Apple's standard bash shell:
5693	@example
5694	ffmpeg -f lavfi -i nullsrc=s=100x100,coreimage=filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
5695	@end example
5696	@end itemize
5697
5698	@section crop
5699
5700	Crop the input video to given dimensions.
5701
5702	It accepts the following parameters:
5703
5704	@table @option
5705	@item w, out_w
5706	The width of the output video. It defaults to @code{iw}.
5707	This expression is evaluated only once during the filter
5708	configuration, or when the @samp{w} or @samp{out_w} command is sent.
5709
5710	@item h, out_h
5711	The height of the output video. It defaults to @code{ih}.
5712	This expression is evaluated only once during the filter
5713	configuration, or when the @samp{h} or @samp{out_h} command is sent.
5714
5715	@item x
5716	The horizontal position, in the input video, of the left edge of the output
5717	video. It defaults to @code{(in_w-out_w)/2}.
5718	This expression is evaluated per-frame.
5719
5720	@item y
5721	The vertical position, in the input video, of the top edge of the output video.
5722	It defaults to @code{(in_h-out_h)/2}.
5723	This expression is evaluated per-frame.
5724
5725	@item keep_aspect
5726	If set to 1 will force the output display aspect ratio
5727	to be the same of the input, by changing the output sample aspect
5728	ratio. It defaults to 0.
5729
5730	@item exact
5731	Enable exact cropping. If enabled, subsampled videos will be cropped at exact
5732	width/height/x/y as specified and will not be rounded to nearest smaller value.
5733	It defaults to 0.
5734	@end table
5735
5736	The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
5737	expressions containing the following constants:
5738
5739	@table @option
5740	@item x
5741	@item y
5742	The computed values for @var{x} and @var{y}. They are evaluated for
5743	each new frame.
5744
5745	@item in_w
5746	@item in_h
5747	The input width and height.
5748
5749	@item iw
5750	@item ih
5751	These are the same as @var{in_w} and @var{in_h}.
5752
5753	@item out_w
5754	@item out_h
5755	The output (cropped) width and height.
5756
5757	@item ow
5758	@item oh
5759	These are the same as @var{out_w} and @var{out_h}.
5760
5761	@item a
5762	same as @var{iw} / @var{ih}
5763
5764	@item sar
5765	input sample aspect ratio
5766
5767	@item dar
5768	input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
5769
5770	@item hsub
5771	@item vsub
5772	horizontal and vertical chroma subsample values. For example for the
5773	pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5774
5775	@item n
5776	The number of the input frame, starting from 0.
5777
5778	@item pos
5779	the position in the file of the input frame, NAN if unknown
5780
5781	@item t
5782	The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
5783
5784	@end table
5785
5786	The expression for @var{out_w} may depend on the value of @var{out_h},
5787	and the expression for @var{out_h} may depend on @var{out_w}, but they
5788	cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
5789	evaluated after @var{out_w} and @var{out_h}.
5790
5791	The @var{x} and @var{y} parameters specify the expressions for the
5792	position of the top-left corner of the output (non-cropped) area. They
5793	are evaluated for each frame. If the evaluated value is not valid, it
5794	is approximated to the nearest valid value.
5795
5796	The expression for @var{x} may depend on @var{y}, and the expression
5797	for @var{y} may depend on @var{x}.
5798
5799	@subsection Examples
5800
5801	@itemize
5802	@item
5803	Crop area with size 100x100 at position (12,34).
5804	@example
5805	crop=100:100:12:34
5806	@end example
5807
5808	Using named options, the example above becomes:
5809	@example
5810	crop=w=100:h=100:x=12:y=34
5811	@end example
5812
5813	@item
5814	Crop the central input area with size 100x100:
5815	@example
5816	crop=100:100
5817	@end example
5818
5819	@item
5820	Crop the central input area with size 2/3 of the input video:
5821	@example
5822	crop=2/3*in_w:2/3*in_h
5823	@end example
5824
5825	@item
5826	Crop the input video central square:
5827	@example
5828	crop=out_w=in_h
5829	crop=in_h
5830	@end example
5831
5832	@item
5833	Delimit the rectangle with the top-left corner placed at position
5834	100:100 and the right-bottom corner corresponding to the right-bottom
5835	corner of the input image.
5836	@example
5837	crop=in_w-100:in_h-100:100:100
5838	@end example
5839
5840	@item
5841	Crop 10 pixels from the left and right borders, and 20 pixels from
5842	the top and bottom borders
5843	@example
5844	crop=in_w-2*10:in_h-2*20
5845	@end example
5846
5847	@item
5848	Keep only the bottom right quarter of the input image:
5849	@example
5850	crop=in_w/2:in_h/2:in_w/2:in_h/2
5851	@end example
5852
5853	@item
5854	Crop height for getting Greek harmony:
5855	@example
5856	crop=in_w:1/PHI*in_w
5857	@end example
5858
5859	@item
5860	Apply trembling effect:
5861	@example
5862	crop=in_w/2:in_h/2:(in_w-out_w)/2+((in_w-out_w)/2)*sin(n/10):(in_h-out_h)/2 +((in_h-out_h)/2)*sin(n/7)
5863	@end example
5864
5865	@item
5866	Apply erratic camera effect depending on timestamp:
5867	@example
5868	crop=in_w/2:in_h/2:(in_w-out_w)/2+((in_w-out_w)/2)*sin(t*10):(in_h-out_h)/2 +((in_h-out_h)/2)*sin(t*13)"
5869	@end example
5870
5871	@item
5872	Set x depending on the value of y:
5873	@example
5874	crop=in_w/2:in_h/2:y:10+10*sin(n/10)
5875	@end example
5876	@end itemize
5877
5878	@subsection Commands
5879
5880	This filter supports the following commands:
5881	@table @option
5882	@item w, out_w
5883	@item h, out_h
5884	@item x
5885	@item y
5886	Set width/height of the output video and the horizontal/vertical position
5887	in the input video.
5888	The command accepts the same syntax of the corresponding option.
5889
5890	If the specified expression is not valid, it is kept at its current
5891	value.
5892	@end table
5893
5894	@section cropdetect
5895
5896	Auto-detect the crop size.
5897
5898	It calculates the necessary cropping parameters and prints the
5899	recommended parameters via the logging system. The detected dimensions
5900	correspond to the non-black area of the input video.
5901
5902	It accepts the following parameters:
5903
5904	@table @option
5905
5906	@item limit
5907	Set higher black value threshold, which can be optionally specified
5908	from nothing (0) to everything (255 for 8-bit based formats). An intensity
5909	value greater to the set value is considered non-black. It defaults to 24.
5910	You can also specify a value between 0.0 and 1.0 which will be scaled depending
5911	on the bitdepth of the pixel format.
5912
5913	@item round
5914	The value which the width/height should be divisible by. It defaults to
5915	16. The offset is automatically adjusted to center the video. Use 2 to
5916	get only even dimensions (needed for 4:2:2 video). 16 is best when
5917	encoding to most video codecs.
5918
5919	@item reset_count, reset
5920	Set the counter that determines after how many frames cropdetect will
5921	reset the previously detected largest video area and start over to
5922	detect the current optimal crop area. Default value is 0.
5923
5924	This can be useful when channel logos distort the video area. 0
5925	indicates 'never reset', and returns the largest area encountered during
5926	playback.
5927	@end table
5928
5929	@anchor{curves}
5930	@section curves
5931
5932	Apply color adjustments using curves.
5933
5934	This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
5935	component (red, green and blue) has its values defined by @var{N} key points
5936	tied from each other using a smooth curve. The x-axis represents the pixel
5937	values from the input frame, and the y-axis the new pixel values to be set for
5938	the output frame.
5939
5940	By default, a component curve is defined by the two points @var{(0;0)} and
5941	@var{(1;1)}. This creates a straight line where each original pixel value is
5942	"adjusted" to its own value, which means no change to the image.
5943
5944	The filter allows you to redefine these two points and add some more. A new
5945	curve (using a natural cubic spline interpolation) will be define to pass
5946	smoothly through all these new coordinates. The new defined points needs to be
5947	strictly increasing over the x-axis, and their @var{x} and @var{y} values must
5948	be in the @var{[0;1]} interval. If the computed curves happened to go outside
5949	the vector spaces, the values will be clipped accordingly.
5950
5951	The filter accepts the following options:
5952
5953	@table @option
5954	@item preset
5955	Select one of the available color presets. This option can be used in addition
5956	to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
5957	options takes priority on the preset values.
5958	Available presets are:
5959	@table @samp
5960	@item none
5961	@item color_negative
5962	@item cross_process
5963	@item darker
5964	@item increase_contrast
5965	@item lighter
5966	@item linear_contrast
5967	@item medium_contrast
5968	@item negative
5969	@item strong_contrast
5970	@item vintage
5971	@end table
5972	Default is @code{none}.
5973	@item master, m
5974	Set the master key points. These points will define a second pass mapping. It
5975	is sometimes called a "luminance" or "value" mapping. It can be used with
5976	@option{r}, @option{g}, @option{b} or @option{all} since it acts like a
5977	post-processing LUT.
5978	@item red, r
5979	Set the key points for the red component.
5980	@item green, g
5981	Set the key points for the green component.
5982	@item blue, b
5983	Set the key points for the blue component.
5984	@item all
5985	Set the key points for all components (not including master).
5986	Can be used in addition to the other key points component
5987	options. In this case, the unset component(s) will fallback on this
5988	@option{all} setting.
5989	@item psfile
5990	Specify a Photoshop curves file (@code{.acv}) to import the settings from.
5991	@item plot
5992	Save Gnuplot script of the curves in specified file.
5993	@end table
5994
5995	To avoid some filtergraph syntax conflicts, each key points list need to be
5996	defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
5997
5998	@subsection Examples
5999
6000	@itemize
6001	@item
6002	Increase slightly the middle level of blue:
6003	@example
6004	curves=blue='0/0 0.5/0.58 1/1'
6005	@end example
6006
6007	@item
6008	Vintage effect:
6009	@example
6010	curves=r='0/0.11 .42/.51 1/0.95':g='0/0 0.50/0.48 1/1':b='0/0.22 .49/.44 1/0.8'
6011	@end example
6012	Here we obtain the following coordinates for each components:
6013	@table @var
6014	@item red
6015	@code{(0;0.11) (0.42;0.51) (1;0.95)}
6016	@item green
6017	@code{(0;0) (0.50;0.48) (1;1)}
6018	@item blue
6019	@code{(0;0.22) (0.49;0.44) (1;0.80)}
6020	@end table
6021
6022	@item
6023	The previous example can also be achieved with the associated built-in preset:
6024	@example
6025	curves=preset=vintage
6026	@end example
6027
6028	@item
6029	Or simply:
6030	@example
6031	curves=vintage
6032	@end example
6033
6034	@item
6035	Use a Photoshop preset and redefine the points of the green component:
6036	@example
6037	curves=psfile='MyCurvesPresets/purple.acv':green='0/0 0.45/0.53 1/1'
6038	@end example
6039
6040	@item
6041	Check out the curves of the @code{cross_process} profile using @command{ffmpeg}
6042	and @command{gnuplot}:
6043	@example
6044	ffmpeg -f lavfi -i color -vf curves=cross_process:plot=/tmp/curves.plt -frames:v 1 -f null -
6045	gnuplot -p /tmp/curves.plt
6046	@end example
6047	@end itemize
6048
6049	@section datascope
6050
6051	Video data analysis filter.
6052
6053	This filter shows hexadecimal pixel values of part of video.
6054
6055	The filter accepts the following options:
6056
6057	@table @option
6058	@item size, s
6059	Set output video size.
6060
6061	@item x
6062	Set x offset from where to pick pixels.
6063
6064	@item y
6065	Set y offset from where to pick pixels.
6066
6067	@item mode
6068	Set scope mode, can be one of the following:
6069	@table @samp
6070	@item mono
6071	Draw hexadecimal pixel values with white color on black background.
6072
6073	@item color
6074	Draw hexadecimal pixel values with input video pixel color on black
6075	background.
6076
6077	@item color2
6078	Draw hexadecimal pixel values on color background picked from input video,
6079	the text color is picked in such way so its always visible.
6080	@end table
6081
6082	@item axis
6083	Draw rows and columns numbers on left and top of video.
6084
6085	@item opacity
6086	Set background opacity.
6087	@end table
6088
6089	@section dctdnoiz
6090
6091	Denoise frames using 2D DCT (frequency domain filtering).
6092
6093	This filter is not designed for real time.
6094
6095	The filter accepts the following options:
6096
6097	@table @option
6098	@item sigma, s
6099	Set the noise sigma constant.
6100
6101	This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
6102	coefficient (absolute value) below this threshold with be dropped.
6103
6104	If you need a more advanced filtering, see @option{expr}.
6105
6106	Default is @code{0}.
6107
6108	@item overlap
6109	Set number overlapping pixels for each block. Since the filter can be slow, you
6110	may want to reduce this value, at the cost of a less effective filter and the
6111	risk of various artefacts.
6112
6113	If the overlapping value doesn't permit processing the whole input width or
6114	height, a warning will be displayed and according borders won't be denoised.
6115
6116	Default value is @var{blocksize}-1, which is the best possible setting.
6117
6118	@item expr, e
6119	Set the coefficient factor expression.
6120
6121	For each coefficient of a DCT block, this expression will be evaluated as a
6122	multiplier value for the coefficient.
6123
6124	If this is option is set, the @option{sigma} option will be ignored.
6125
6126	The absolute value of the coefficient can be accessed through the @var{c}
6127	variable.
6128
6129	@item n
6130	Set the @var{blocksize} using the number of bits. @code{1<<@var{n}} defines the
6131	@var{blocksize}, which is the width and height of the processed blocks.
6132
6133	The default value is @var{3} (8x8) and can be raised to @var{4} for a
6134	@var{blocksize} of 16x16. Note that changing this setting has huge consequences
6135	on the speed processing. Also, a larger block size does not necessarily means a
6136	better de-noising.
6137	@end table
6138
6139	@subsection Examples
6140
6141	Apply a denoise with a @option{sigma} of @code{4.5}:
6142	@example
6143	dctdnoiz=4.5
6144	@end example
6145
6146	The same operation can be achieved using the expression system:
6147	@example
6148	dctdnoiz=e='gte(c, 4.5*3)'
6149	@end example
6150
6151	Violent denoise using a block size of @code{16x16}:
6152	@example
6153	dctdnoiz=15:n=4
6154	@end example
6155
6156	@section deband
6157
6158	Remove banding artifacts from input video.
6159	It works by replacing banded pixels with average value of referenced pixels.
6160
6161	The filter accepts the following options:
6162
6163	@table @option
6164	@item 1thr
6165	@item 2thr
6166	@item 3thr
6167	@item 4thr
6168	Set banding detection threshold for each plane. Default is 0.02.
6169	Valid range is 0.00003 to 0.5.
6170	If difference between current pixel and reference pixel is less than threshold,
6171	it will be considered as banded.
6172
6173	@item range, r
6174	Banding detection range in pixels. Default is 16. If positive, random number
6175	in range 0 to set value will be used. If negative, exact absolute value
6176	will be used.
6177	The range defines square of four pixels around current pixel.
6178
6179	@item direction, d
6180	Set direction in radians from which four pixel will be compared. If positive,
6181	random direction from 0 to set direction will be picked. If negative, exact of
6182	absolute value will be picked. For example direction 0, -PI or -2*PI radians
6183	will pick only pixels on same row and -PI/2 will pick only pixels on same
6184	column.
6185
6186	@item blur, b
6187	If enabled, current pixel is compared with average value of all four
6188	surrounding pixels. The default is enabled. If disabled current pixel is
6189	compared with all four surrounding pixels. The pixel is considered banded
6190	if only all four differences with surrounding pixels are less than threshold.
6191
6192	@item coupling, c
6193	If enabled, current pixel is changed if and only if all pixel components are banded,
6194	e.g. banding detection threshold is triggered for all color components.
6195	The default is disabled.
6196	@end table
6197
6198	@anchor{decimate}
6199	@section decimate
6200
6201	Drop duplicated frames at regular intervals.
6202
6203	The filter accepts the following options:
6204
6205	@table @option
6206	@item cycle
6207	Set the number of frames from which one will be dropped. Setting this to
6208	@var{N} means one frame in every batch of @var{N} frames will be dropped.
6209	Default is @code{5}.
6210
6211	@item dupthresh
6212	Set the threshold for duplicate detection. If the difference metric for a frame
6213	is less than or equal to this value, then it is declared as duplicate. Default
6214	is @code{1.1}
6215
6216	@item scthresh
6217	Set scene change threshold. Default is @code{15}.
6218
6219	@item blockx
6220	@item blocky
6221	Set the size of the x and y-axis blocks used during metric calculations.
6222	Larger blocks give better noise suppression, but also give worse detection of
6223	small movements. Must be a power of two. Default is @code{32}.
6224
6225	@item ppsrc
6226	Mark main input as a pre-processed input and activate clean source input
6227	stream. This allows the input to be pre-processed with various filters to help
6228	the metrics calculation while keeping the frame selection lossless. When set to
6229	@code{1}, the first stream is for the pre-processed input, and the second
6230	stream is the clean source from where the kept frames are chosen. Default is
6231	@code{0}.
6232
6233	@item chroma
6234	Set whether or not chroma is considered in the metric calculations. Default is
6235	@code{1}.
6236	@end table
6237
6238	@section deflate
6239
6240	Apply deflate effect to the video.
6241
6242	This filter replaces the pixel by the local(3x3) average by taking into account
6243	only values lower than the pixel.
6244
6245	It accepts the following options:
6246
6247	@table @option
6248	@item threshold0
6249	@item threshold1
6250	@item threshold2
6251	@item threshold3
6252	Limit the maximum change for each plane, default is 65535.
6253	If 0, plane will remain unchanged.
6254	@end table
6255
6256	@section dejudder
6257
6258	Remove judder produced by partially interlaced telecined content.
6259
6260	Judder can be introduced, for instance, by @ref{pullup} filter. If the original
6261	source was partially telecined content then the output of @code{pullup,dejudder}
6262	will have a variable frame rate. May change the recorded frame rate of the
6263	container. Aside from that change, this filter will not affect constant frame
6264	rate video.
6265
6266	The option available in this filter is:
6267	@table @option
6268
6269	@item cycle
6270	Specify the length of the window over which the judder repeats.
6271
6272	Accepts any integer greater than 1. Useful values are:
6273	@table @samp
6274
6275	@item 4
6276	If the original was telecined from 24 to 30 fps (Film to NTSC).
6277
6278	@item 5
6279	If the original was telecined from 25 to 30 fps (PAL to NTSC).
6280
6281	@item 20
6282	If a mixture of the two.
6283	@end table
6284
6285	The default is @samp{4}.
6286	@end table
6287
6288	@section delogo
6289
6290	Suppress a TV station logo by a simple interpolation of the surrounding
6291	pixels. Just set a rectangle covering the logo and watch it disappear
6292	(and sometimes something even uglier appear - your mileage may vary).
6293
6294	It accepts the following parameters:
6295	@table @option
6296
6297	@item x
6298	@item y
6299	Specify the top left corner coordinates of the logo. They must be
6300	specified.
6301
6302	@item w
6303	@item h
6304	Specify the width and height of the logo to clear. They must be
6305	specified.
6306
6307	@item band, t
6308	Specify the thickness of the fuzzy edge of the rectangle (added to
6309	@var{w} and @var{h}). The default value is 1. This option is
6310	deprecated, setting higher values should no longer be necessary and
6311	is not recommended.
6312
6313	@item show
6314	When set to 1, a green rectangle is drawn on the screen to simplify
6315	finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
6316	The default value is 0.
6317
6318	The rectangle is drawn on the outermost pixels which will be (partly)
6319	replaced with interpolated values. The values of the next pixels
6320	immediately outside this rectangle in each direction will be used to
6321	compute the interpolated pixel values inside the rectangle.
6322
6323	@end table
6324
6325	@subsection Examples
6326
6327	@itemize
6328	@item
6329	Set a rectangle covering the area with top left corner coordinates 0,0
6330	and size 100x77, and a band of size 10:
6331	@example
6332	delogo=x=0:y=0:w=100:h=77:band=10
6333	@end example
6334
6335	@end itemize
6336
6337	@section deshake
6338
6339	Attempt to fix small changes in horizontal and/or vertical shift. This
6340	filter helps remove camera shake from hand-holding a camera, bumping a
6341	tripod, moving on a vehicle, etc.
6342
6343	The filter accepts the following options:
6344
6345	@table @option
6346
6347	@item x
6348	@item y
6349	@item w
6350	@item h
6351	Specify a rectangular area where to limit the search for motion
6352	vectors.
6353	If desired the search for motion vectors can be limited to a
6354	rectangular area of the frame defined by its top left corner, width
6355	and height. These parameters have the same meaning as the drawbox
6356	filter which can be used to visualise the position of the bounding
6357	box.
6358
6359	This is useful when simultaneous movement of subjects within the frame
6360	might be confused for camera motion by the motion vector search.
6361
6362	If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
6363	then the full frame is used. This allows later options to be set
6364	without specifying the bounding box for the motion vector search.
6365
6366	Default - search the whole frame.
6367
6368	@item rx
6369	@item ry
6370	Specify the maximum extent of movement in x and y directions in the
6371	range 0-64 pixels. Default 16.
6372
6373	@item edge
6374	Specify how to generate pixels to fill blanks at the edge of the
6375	frame. Available values are:
6376	@table @samp
6377	@item blank, 0
6378	Fill zeroes at blank locations
6379	@item original, 1
6380	Original image at blank locations
6381	@item clamp, 2
6382	Extruded edge value at blank locations
6383	@item mirror, 3
6384	Mirrored edge at blank locations
6385	@end table
6386	Default value is @samp{mirror}.
6387
6388	@item blocksize
6389	Specify the blocksize to use for motion search. Range 4-128 pixels,
6390	default 8.
6391
6392	@item contrast
6393	Specify the contrast threshold for blocks. Only blocks with more than
6394	the specified contrast (difference between darkest and lightest
6395	pixels) will be considered. Range 1-255, default 125.
6396
6397	@item search
6398	Specify the search strategy. Available values are:
6399	@table @samp
6400	@item exhaustive, 0
6401	Set exhaustive search
6402	@item less, 1
6403	Set less exhaustive search.
6404	@end table
6405	Default value is @samp{exhaustive}.
6406
6407	@item filename
6408	If set then a detailed log of the motion search is written to the
6409	specified file.
6410
6411	@item opencl
6412	If set to 1, specify using OpenCL capabilities, only available if
6413	FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
6414
6415	@end table
6416
6417	@section detelecine
6418
6419	Apply an exact inverse of the telecine operation. It requires a predefined
6420	pattern specified using the pattern option which must be the same as that passed
6421	to the telecine filter.
6422
6423	This filter accepts the following options:
6424
6425	@table @option
6426	@item first_field
6427	@table @samp
6428	@item top, t
6429	top field first
6430	@item bottom, b
6431	bottom field first
6432	The default value is @code{top}.
6433	@end table
6434
6435	@item pattern
6436	A string of numbers representing the pulldown pattern you wish to apply.
6437	The default value is @code{23}.
6438
6439	@item start_frame
6440	A number representing position of the first frame with respect to the telecine
6441	pattern. This is to be used if the stream is cut. The default value is @code{0}.
6442	@end table
6443
6444	@section dilation
6445
6446	Apply dilation effect to the video.
6447
6448	This filter replaces the pixel by the local(3x3) maximum.
6449
6450	It accepts the following options:
6451
6452	@table @option
6453	@item threshold0
6454	@item threshold1
6455	@item threshold2
6456	@item threshold3
6457	Limit the maximum change for each plane, default is 65535.
6458	If 0, plane will remain unchanged.
6459
6460	@item coordinates
6461	Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
6462	pixels are used.
6463
6464	Flags to local 3x3 coordinates maps like this:
6465
6466	1 2 3
6467	4 5
6468	6 7 8
6469	@end table
6470
6471	@section displace
6472
6473	Displace pixels as indicated by second and third input stream.
6474
6475	It takes three input streams and outputs one stream, the first input is the
6476	source, and second and third input are displacement maps.
6477
6478	The second input specifies how much to displace pixels along the
6479	x-axis, while the third input specifies how much to displace pixels
6480	along the y-axis.
6481	If one of displacement map streams terminates, last frame from that
6482	displacement map will be used.
6483
6484	Note that once generated, displacements maps can be reused over and over again.
6485
6486	A description of the accepted options follows.
6487
6488	@table @option
6489	@item edge
6490	Set displace behavior for pixels that are out of range.
6491
6492	Available values are:
6493	@table @samp
6494	@item blank
6495	Missing pixels are replaced by black pixels.
6496
6497	@item smear
6498	Adjacent pixels will spread out to replace missing pixels.
6499
6500	@item wrap
6501	Out of range pixels are wrapped so they point to pixels of other side.
6502	@end table
6503	Default is @samp{smear}.
6504
6505	@end table
6506
6507	@subsection Examples
6508
6509	@itemize
6510	@item
6511	Add ripple effect to rgb input of video size hd720:
6512	@example
6513	ffmpeg -i INPUT -f lavfi -i nullsrc=s=hd720,lutrgb=128:128:128 -f lavfi -i nullsrc=s=hd720,geq='r=128+30*sin(2*PI*X/400+T):g=128+30*sin(2*PI*X/400+T):b=128+30*sin(2*PI*X/400+T)' -lavfi '[0][1][2]displace' OUTPUT
6514	@end example
6515
6516	@item
6517	Add wave effect to rgb input of video size hd720:
6518	@example
6519	ffmpeg -i INPUT -f lavfi -i nullsrc=hd720,geq='r=128+80*(sin(sqrt((X-W/2)*(X-W/2)+(Y-H/2)*(Y-H/2))/220*2*PI+T)):g=128+80*(sin(sqrt((X-W/2)*(X-W/2)+(Y-H/2)*(Y-H/2))/220*2*PI+T)):b=128+80*(sin(sqrt((X-W/2)*(X-W/2)+(Y-H/2)*(Y-H/2))/220*2*PI+T))' -lavfi '[1]split[x][y],[0][x][y]displace' OUTPUT
6520	@end example
6521	@end itemize
6522
6523	@section drawbox
6524
6525	Draw a colored box on the input image.
6526
6527	It accepts the following parameters:
6528
6529	@table @option
6530	@item x
6531	@item y
6532	The expressions which specify the top left corner coordinates of the box. It defaults to 0.
6533
6534	@item width, w
6535	@item height, h
6536	The expressions which specify the width and height of the box; if 0 they are interpreted as
6537	the input width and height. It defaults to 0.
6538
6539	@item color, c
6540	Specify the color of the box to write. For the general syntax of this option,
6541	check the "Color" section in the ffmpeg-utils manual. If the special
6542	value @code{invert} is used, the box edge color is the same as the
6543	video with inverted luma.
6544
6545	@item thickness, t
6546	The expression which sets the thickness of the box edge. Default value is @code{3}.
6547
6548	See below for the list of accepted constants.
6549	@end table
6550
6551	The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6552	following constants:
6553
6554	@table @option
6555	@item dar
6556	The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6557
6558	@item hsub
6559	@item vsub
6560	horizontal and vertical chroma subsample values. For example for the
6561	pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6562
6563	@item in_h, ih
6564	@item in_w, iw
6565	The input width and height.
6566
6567	@item sar
6568	The input sample aspect ratio.
6569
6570	@item x
6571	@item y
6572	The x and y offset coordinates where the box is drawn.
6573
6574	@item w
6575	@item h
6576	The width and height of the drawn box.
6577
6578	@item t
6579	The thickness of the drawn box.
6580
6581	These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6582	each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6583
6584	@end table
6585
6586	@subsection Examples
6587
6588	@itemize
6589	@item
6590	Draw a black box around the edge of the input image:
6591	@example
6592	drawbox
6593	@end example
6594
6595	@item
6596	Draw a box with color red and an opacity of 50%:
6597	@example
6598	drawbox=10:20:200:60:red@@0.5
6599	@end example
6600
6601	The previous example can be specified as:
6602	@example
6603	drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
6604	@end example
6605
6606	@item
6607	Fill the box with pink color:
6608	@example
6609	drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
6610	@end example
6611
6612	@item
6613	Draw a 2-pixel red 2.40:1 mask:
6614	@example
6615	drawbox=x=-t:y=0.5*(ih-iw/2.4)-t:w=iw+t*2:h=iw/2.4+t*2:t=2:c=red
6616	@end example
6617	@end itemize
6618
6619	@section drawgrid
6620
6621	Draw a grid on the input image.
6622
6623	It accepts the following parameters:
6624
6625	@table @option
6626	@item x
6627	@item y
6628	The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
6629
6630	@item width, w
6631	@item height, h
6632	The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
6633	input width and height, respectively, minus @code{thickness}, so image gets
6634	framed. Default to 0.
6635
6636	@item color, c
6637	Specify the color of the grid. For the general syntax of this option,
6638	check the "Color" section in the ffmpeg-utils manual. If the special
6639	value @code{invert} is used, the grid color is the same as the
6640	video with inverted luma.
6641
6642	@item thickness, t
6643	The expression which sets the thickness of the grid line. Default value is @code{1}.
6644
6645	See below for the list of accepted constants.
6646	@end table
6647
6648	The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6649	following constants:
6650
6651	@table @option
6652	@item dar
6653	The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6654
6655	@item hsub
6656	@item vsub
6657	horizontal and vertical chroma subsample values. For example for the
6658	pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6659
6660	@item in_h, ih
6661	@item in_w, iw
6662	The input grid cell width and height.
6663
6664	@item sar
6665	The input sample aspect ratio.
6666
6667	@item x
6668	@item y
6669	The x and y coordinates of some point of grid intersection (meant to configure offset).
6670
6671	@item w
6672	@item h
6673	The width and height of the drawn cell.
6674
6675	@item t
6676	The thickness of the drawn cell.
6677
6678	These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6679	each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6680
6681	@end table
6682
6683	@subsection Examples
6684
6685	@itemize
6686	@item
6687	Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
6688	@example
6689	drawgrid=width=100:height=100:thickness=2:color=red@@0.5
6690	@end example
6691
6692	@item
6693	Draw a white 3x3 grid with an opacity of 50%:
6694	@example
6695	drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
6696	@end example
6697	@end itemize
6698
6699	@anchor{drawtext}
6700	@section drawtext
6701
6702	Draw a text string or text from a specified file on top of a video, using the
6703	libfreetype library.
6704
6705	To enable compilation of this filter, you need to configure FFmpeg with
6706	@code{--enable-libfreetype}.
6707	To enable default font fallback and the @var{font} option you need to
6708	configure FFmpeg with @code{--enable-libfontconfig}.
6709	To enable the @var{text_shaping} option, you need to configure FFmpeg with
6710	@code{--enable-libfribidi}.
6711
6712	@subsection Syntax
6713
6714	It accepts the following parameters:
6715
6716	@table @option
6717
6718	@item box
6719	Used to draw a box around text using the background color.
6720	The value must be either 1 (enable) or 0 (disable).
6721	The default value of @var{box} is 0.
6722
6723	@item boxborderw
6724	Set the width of the border to be drawn around the box using @var{boxcolor}.
6725	The default value of @var{boxborderw} is 0.
6726
6727	@item boxcolor
6728	The color to be used for drawing box around text. For the syntax of this
6729	option, check the "Color" section in the ffmpeg-utils manual.
6730
6731	The default value of @var{boxcolor} is "white".
6732
6733	@item line_spacing
6734	Set the line spacing in pixels of the border to be drawn around the box using @var{box}.
6735	The default value of @var{line_spacing} is 0.
6736
6737	@item borderw
6738	Set the width of the border to be drawn around the text using @var{bordercolor}.
6739	The default value of @var{borderw} is 0.
6740
6741	@item bordercolor
6742	Set the color to be used for drawing border around text. For the syntax of this
6743	option, check the "Color" section in the ffmpeg-utils manual.
6744
6745	The default value of @var{bordercolor} is "black".
6746
6747	@item expansion
6748	Select how the @var{text} is expanded. Can be either @code{none},
6749	@code{strftime} (deprecated) or
6750	@code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
6751	below for details.
6752
6753	@item basetime
6754	Set a start time for the count. Value is in microseconds. Only applied
6755	in the deprecated strftime expansion mode. To emulate in normal expansion
6756	mode use the @code{pts} function, supplying the start time (in seconds)
6757	as the second argument.
6758
6759	@item fix_bounds
6760	If true, check and fix text coords to avoid clipping.
6761
6762	@item fontcolor
6763	The color to be used for drawing fonts. For the syntax of this option, check
6764	the "Color" section in the ffmpeg-utils manual.
6765
6766	The default value of @var{fontcolor} is "black".
6767
6768	@item fontcolor_expr
6769	String which is expanded the same way as @var{text} to obtain dynamic
6770	@var{fontcolor} value. By default this option has empty value and is not
6771	processed. When this option is set, it overrides @var{fontcolor} option.
6772
6773	@item font
6774	The font family to be used for drawing text. By default Sans.
6775
6776	@item fontfile
6777	The font file to be used for drawing text. The path must be included.
6778	This parameter is mandatory if the fontconfig support is disabled.
6779
6780	@item alpha
6781	Draw the text applying alpha blending. The value can
6782	be a number between 0.0 and 1.0.
6783	The expression accepts the same variables @var{x, y} as well.
6784	The default value is 1.
6785	Please see @var{fontcolor_expr}.
6786
6787	@item fontsize
6788	The font size to be used for drawing text.
6789	The default value of @var{fontsize} is 16.
6790
6791	@item text_shaping
6792	If set to 1, attempt to shape the text (for example, reverse the order of
6793	right-to-left text and join Arabic characters) before drawing it.
6794	Otherwise, just draw the text exactly as given.
6795	By default 1 (if supported).
6796
6797	@item ft_load_flags
6798	The flags to be used for loading the fonts.
6799
6800	The flags map the corresponding flags supported by libfreetype, and are
6801	a combination of the following values:
6802	@table @var
6803	@item default
6804	@item no_scale
6805	@item no_hinting
6806	@item render
6807	@item no_bitmap
6808	@item vertical_layout
6809	@item force_autohint
6810	@item crop_bitmap
6811	@item pedantic
6812	@item ignore_global_advance_width
6813	@item no_recurse
6814	@item ignore_transform
6815	@item monochrome
6816	@item linear_design
6817	@item no_autohint
6818	@end table
6819
6820	Default value is "default".
6821
6822	For more information consult the documentation for the FT_LOAD_*
6823	libfreetype flags.
6824
6825	@item shadowcolor
6826	The color to be used for drawing a shadow behind the drawn text. For the
6827	syntax of this option, check the "Color" section in the ffmpeg-utils manual.
6828
6829	The default value of @var{shadowcolor} is "black".
6830
6831	@item shadowx
6832	@item shadowy
6833	The x and y offsets for the text shadow position with respect to the
6834	position of the text. They can be either positive or negative
6835	values. The default value for both is "0".
6836
6837	@item start_number
6838	The starting frame number for the n/frame_num variable. The default value
6839	is "0".
6840
6841	@item tabsize
6842	The size in number of spaces to use for rendering the tab.
6843	Default value is 4.
6844
6845	@item timecode
6846	Set the initial timecode representation in "hh:mm:ss[:;.]ff"
6847	format. It can be used with or without text parameter. @var{timecode_rate}
6848	option must be specified.
6849
6850	@item timecode_rate, rate, r
6851	Set the timecode frame rate (timecode only).
6852
6853	@item tc24hmax
6854	If set to 1, the output of the timecode option will wrap around at 24 hours.
6855	Default is 0 (disabled).
6856
6857	@item text
6858	The text string to be drawn. The text must be a sequence of UTF-8
6859	encoded characters.
6860	This parameter is mandatory if no file is specified with the parameter
6861	@var{textfile}.
6862
6863	@item textfile
6864	A text file containing text to be drawn. The text must be a sequence
6865	of UTF-8 encoded characters.
6866
6867	This parameter is mandatory if no text string is specified with the
6868	parameter @var{text}.
6869
6870	If both @var{text} and @var{textfile} are specified, an error is thrown.
6871
6872	@item reload
6873	If set to 1, the @var{textfile} will be reloaded before each frame.
6874	Be sure to update it atomically, or it may be read partially, or even fail.
6875
6876	@item x
6877	@item y
6878	The expressions which specify the offsets where text will be drawn
6879	within the video frame. They are relative to the top/left border of the
6880	output image.
6881
6882	The default value of @var{x} and @var{y} is "0".
6883
6884	See below for the list of accepted constants and functions.
6885	@end table
6886
6887	The parameters for @var{x} and @var{y} are expressions containing the
6888	following constants and functions:
6889
6890	@table @option
6891	@item dar
6892	input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
6893
6894	@item hsub
6895	@item vsub
6896	horizontal and vertical chroma subsample values. For example for the
6897	pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6898
6899	@item line_h, lh
6900	the height of each text line
6901
6902	@item main_h, h, H
6903	the input height
6904
6905	@item main_w, w, W
6906	the input width
6907
6908	@item max_glyph_a, ascent
6909	the maximum distance from the baseline to the highest/upper grid
6910	coordinate used to place a glyph outline point, for all the rendered
6911	glyphs.
6912	It is a positive value, due to the grid's orientation with the Y axis
6913	upwards.
6914
6915	@item max_glyph_d, descent
6916	the maximum distance from the baseline to the lowest grid coordinate
6917	used to place a glyph outline point, for all the rendered glyphs.
6918	This is a negative value, due to the grid's orientation, with the Y axis
6919	upwards.
6920
6921	@item max_glyph_h
6922	maximum glyph height, that is the maximum height for all the glyphs
6923	contained in the rendered text, it is equivalent to @var{ascent} -
6924	@var{descent}.
6925
6926	@item max_glyph_w
6927	maximum glyph width, that is the maximum width for all the glyphs
6928	contained in the rendered text
6929
6930	@item n
6931	the number of input frame, starting from 0
6932
6933	@item rand(min, max)
6934	return a random number included between @var{min} and @var{max}
6935
6936	@item sar
6937	The input sample aspect ratio.
6938
6939	@item t
6940	timestamp expressed in seconds, NAN if the input timestamp is unknown
6941
6942	@item text_h, th
6943	the height of the rendered text
6944
6945	@item text_w, tw
6946	the width of the rendered text
6947
6948	@item x
6949	@item y
6950	the x and y offset coordinates where the text is drawn.
6951
6952	These parameters allow the @var{x} and @var{y} expressions to refer
6953	each other, so you can for example specify @code{y=x/dar}.
6954	@end table
6955
6956	@anchor{drawtext_expansion}
6957	@subsection Text expansion
6958
6959	If @option{expansion} is set to @code{strftime},
6960	the filter recognizes strftime() sequences in the provided text and
6961	expands them accordingly. Check the documentation of strftime(). This
6962	feature is deprecated.
6963
6964	If @option{expansion} is set to @code{none}, the text is printed verbatim.
6965
6966	If @option{expansion} is set to @code{normal} (which is the default),
6967	the following expansion mechanism is used.
6968
6969	The backslash character @samp{\}, followed by any character, always expands to
6970	the second character.
6971
6972	Sequences of the form @code{%@{...@}} are expanded. The text between the
6973	braces is a function name, possibly followed by arguments separated by ':'.
6974	If the arguments contain special characters or delimiters (':' or '@}'),
6975	they should be escaped.
6976
6977	Note that they probably must also be escaped as the value for the
6978	@option{text} option in the filter argument string and as the filter
6979	argument in the filtergraph description, and possibly also for the shell,
6980	that makes up to four levels of escaping; using a text file avoids these
6981	problems.
6982
6983	The following functions are available:
6984
6985	@table @command
6986
6987	@item expr, e
6988	The expression evaluation result.
6989
6990	It must take one argument specifying the expression to be evaluated,
6991	which accepts the same constants and functions as the @var{x} and
6992	@var{y} values. Note that not all constants should be used, for
6993	example the text size is not known when evaluating the expression, so
6994	the constants @var{text_w} and @var{text_h} will have an undefined
6995	value.
6996
6997	@item expr_int_format, eif
6998	Evaluate the expression's value and output as formatted integer.
6999
7000	The first argument is the expression to be evaluated, just as for the @var{expr} function.
7001	The second argument specifies the output format. Allowed values are @samp{x},
7002	@samp{X}, @samp{d} and @samp{u}. They are treated exactly as in the
7003	@code{printf} function.
7004	The third parameter is optional and sets the number of positions taken by the output.
7005	It can be used to add padding with zeros from the left.
7006
7007	@item gmtime
7008	The time at which the filter is running, expressed in UTC.
7009	It can accept an argument: a strftime() format string.
7010
7011	@item localtime
7012	The time at which the filter is running, expressed in the local time zone.
7013	It can accept an argument: a strftime() format string.
7014
7015	@item metadata
7016	Frame metadata. Takes one or two arguments.
7017
7018	The first argument is mandatory and specifies the metadata key.
7019
7020	The second argument is optional and specifies a default value, used when the
7021	metadata key is not found or empty.
7022
7023	@item n, frame_num
7024	The frame number, starting from 0.
7025
7026	@item pict_type
7027	A 1 character description of the current picture type.
7028
7029	@item pts
7030	The timestamp of the current frame.
7031	It can take up to three arguments.
7032
7033	The first argument is the format of the timestamp; it defaults to @code{flt}
7034	for seconds as a decimal number with microsecond accuracy; @code{hms} stands
7035	for a formatted @var{[-]HH:MM:SS.mmm} timestamp with millisecond accuracy.
7036	@code{gmtime} stands for the timestamp of the frame formatted as UTC time;
7037	@code{localtime} stands for the timestamp of the frame formatted as
7038	local time zone time.
7039
7040	The second argument is an offset added to the timestamp.
7041
7042	If the format is set to @code{localtime} or @code{gmtime},
7043	a third argument may be supplied: a strftime() format string.
7044	By default, @var{YYYY-MM-DD HH:MM:SS} format will be used.
7045	@end table
7046
7047	@subsection Examples
7048
7049	@itemize
7050	@item
7051	Draw "Test Text" with font FreeSerif, using the default values for the
7052	optional parameters.
7053
7054	@example
7055	drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
7056	@end example
7057
7058	@item
7059	Draw 'Test Text' with font FreeSerif of size 24 at position x=100
7060	and y=50 (counting from the top-left corner of the screen), text is
7061	yellow with a red box around it. Both the text and the box have an
7062	opacity of 20%.
7063
7064	@example
7065	drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
7066	x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
7067	@end example
7068
7069	Note that the double quotes are not necessary if spaces are not used
7070	within the parameter list.
7071
7072	@item
7073	Show the text at the center of the video frame:
7074	@example
7075	drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h)/2"
7076	@end example
7077
7078	@item
7079	Show the text at a random position, switching to a new position every 30 seconds:
7080	@example
7081	drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=if(eq(mod(t\,30)\,0)\,rand(0\,(w-text_w))\,x):y=if(eq(mod(t\,30)\,0)\,rand(0\,(h-text_h))\,y)"
7082	@end example
7083
7084	@item
7085	Show a text line sliding from right to left in the last row of the video
7086	frame. The file @file{LONG_LINE} is assumed to contain a single line
7087	with no newlines.
7088	@example
7089	drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
7090	@end example
7091
7092	@item
7093	Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
7094	@example
7095	drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
7096	@end example
7097
7098	@item
7099	Draw a single green letter "g", at the center of the input video.
7100	The glyph baseline is placed at half screen height.
7101	@example
7102	drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
7103	@end example
7104
7105	@item
7106	Show text for 1 second every 3 seconds:
7107	@example
7108	drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
7109	@end example
7110
7111	@item
7112	Use fontconfig to set the font. Note that the colons need to be escaped.
7113	@example
7114	drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
7115	@end example
7116
7117	@item
7118	Print the date of a real-time encoding (see strftime(3)):
7119	@example
7120	drawtext='fontfile=FreeSans.ttf:text=%@{localtime\:%a %b %d %Y@}'
7121	@end example
7122
7123	@item
7124	Show text fading in and out (appearing/disappearing):
7125	@example
7126	#!/bin/sh
7127	DS=1.0 # display start
7128	DE=10.0 # display end
7129	FID=1.5 # fade in duration
7130	FOD=5 # fade out duration
7131	ffplay -f lavfi "color,drawtext=text=TEST:fontsize=50:fontfile=FreeSerif.ttf:fontcolor_expr=ff0000%@{eif\\\\: clip(255*(1*between(t\\, $DS + $FID\\, $DE - $FOD) + ((t - $DS)/$FID)*between(t\\, $DS\\, $DS + $FID) + (-(t - $DE)/$FOD)*between(t\\, $DE - $FOD\\, $DE) )\\, 0\\, 255) \\\\: x\\\\: 2 @}"
7132	@end example
7133
7134	@item
7135	Horizontally align multiple separate texts. Note that @option{max_glyph_a}
7136	and the @option{fontsize} value are included in the @option{y} offset.
7137	@example
7138	drawtext=fontfile=FreeSans.ttf:text=DOG:fontsize=24:x=10:y=20+24-max_glyph_a,
7139	drawtext=fontfile=FreeSans.ttf:text=cow:fontsize=24:x=80:y=20+24-max_glyph_a
7140	@end example
7141
7142	@end itemize
7143
7144	For more information about libfreetype, check:
7145	@url{http://www.freetype.org/}.
7146
7147	For more information about fontconfig, check:
7148	@url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
7149
7150	For more information about libfribidi, check:
7151	@url{http://fribidi.org/}.
7152
7153	@section edgedetect
7154
7155	Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
7156
7157	The filter accepts the following options:
7158
7159	@table @option
7160	@item low
7161	@item high
7162	Set low and high threshold values used by the Canny thresholding
7163	algorithm.
7164
7165	The high threshold selects the "strong" edge pixels, which are then
7166	connected through 8-connectivity with the "weak" edge pixels selected
7167	by the low threshold.
7168
7169	@var{low} and @var{high} threshold values must be chosen in the range
7170	[0,1], and @var{low} should be lesser or equal to @var{high}.
7171
7172	Default value for @var{low} is @code{20/255}, and default value for @var{high}
7173	is @code{50/255}.
7174
7175	@item mode
7176	Define the drawing mode.
7177
7178	@table @samp
7179	@item wires
7180	Draw white/gray wires on black background.
7181
7182	@item colormix
7183	Mix the colors to create a paint/cartoon effect.
7184	@end table
7185
7186	Default value is @var{wires}.
7187	@end table
7188
7189	@subsection Examples
7190
7191	@itemize
7192	@item
7193	Standard edge detection with custom values for the hysteresis thresholding:
7194	@example
7195	edgedetect=low=0.1:high=0.4
7196	@end example
7197
7198	@item
7199	Painting effect without thresholding:
7200	@example
7201	edgedetect=mode=colormix:high=0
7202	@end example
7203	@end itemize
7204
7205	@section eq
7206	Set brightness, contrast, saturation and approximate gamma adjustment.
7207
7208	The filter accepts the following options:
7209
7210	@table @option
7211	@item contrast
7212	Set the contrast expression. The value must be a float value in range
7213	@code{-2.0} to @code{2.0}. The default value is "1".
7214
7215	@item brightness
7216	Set the brightness expression. The value must be a float value in
7217	range @code{-1.0} to @code{1.0}. The default value is "0".
7218
7219	@item saturation
7220	Set the saturation expression. The value must be a float in
7221	range @code{0.0} to @code{3.0}. The default value is "1".
7222
7223	@item gamma
7224	Set the gamma expression. The value must be a float in range
7225	@code{0.1} to @code{10.0}. The default value is "1".
7226
7227	@item gamma_r
7228	Set the gamma expression for red. The value must be a float in
7229	range @code{0.1} to @code{10.0}. The default value is "1".
7230
7231	@item gamma_g
7232	Set the gamma expression for green. The value must be a float in range
7233	@code{0.1} to @code{10.0}. The default value is "1".
7234
7235	@item gamma_b
7236	Set the gamma expression for blue. The value must be a float in range
7237	@code{0.1} to @code{10.0}. The default value is "1".
7238
7239	@item gamma_weight
7240	Set the gamma weight expression. It can be used to reduce the effect
7241	of a high gamma value on bright image areas, e.g. keep them from
7242	getting overamplified and just plain white. The value must be a float
7243	in range @code{0.0} to @code{1.0}. A value of @code{0.0} turns the
7244	gamma correction all the way down while @code{1.0} leaves it at its
7245	full strength. Default is "1".
7246
7247	@item eval
7248	Set when the expressions for brightness, contrast, saturation and
7249	gamma expressions are evaluated.
7250
7251	It accepts the following values:
7252	@table @samp
7253	@item init
7254	only evaluate expressions once during the filter initialization or
7255	when a command is processed
7256
7257	@item frame
7258	evaluate expressions for each incoming frame
7259	@end table
7260
7261	Default value is @samp{init}.
7262	@end table
7263
7264	The expressions accept the following parameters:
7265	@table @option
7266	@item n
7267	frame count of the input frame starting from 0
7268
7269	@item pos
7270	byte position of the corresponding packet in the input file, NAN if
7271	unspecified
7272
7273	@item r
7274	frame rate of the input video, NAN if the input frame rate is unknown
7275
7276	@item t
7277	timestamp expressed in seconds, NAN if the input timestamp is unknown
7278	@end table
7279
7280	@subsection Commands
7281	The filter supports the following commands:
7282
7283	@table @option
7284	@item contrast
7285	Set the contrast expression.
7286
7287	@item brightness
7288	Set the brightness expression.
7289
7290	@item saturation
7291	Set the saturation expression.
7292
7293	@item gamma
7294	Set the gamma expression.
7295
7296	@item gamma_r
7297	Set the gamma_r expression.
7298
7299	@item gamma_g
7300	Set gamma_g expression.
7301
7302	@item gamma_b
7303	Set gamma_b expression.
7304
7305	@item gamma_weight
7306	Set gamma_weight expression.
7307
7308	The command accepts the same syntax of the corresponding option.
7309
7310	If the specified expression is not valid, it is kept at its current
7311	value.
7312
7313	@end table
7314
7315	@section erosion
7316
7317	Apply erosion effect to the video.
7318
7319	This filter replaces the pixel by the local(3x3) minimum.
7320
7321	It accepts the following options:
7322
7323	@table @option
7324	@item threshold0
7325	@item threshold1
7326	@item threshold2
7327	@item threshold3
7328	Limit the maximum change for each plane, default is 65535.
7329	If 0, plane will remain unchanged.
7330
7331	@item coordinates
7332	Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
7333	pixels are used.
7334
7335	Flags to local 3x3 coordinates maps like this:
7336
7337	1 2 3
7338	4 5
7339	6 7 8
7340	@end table
7341
7342	@section extractplanes
7343
7344	Extract color channel components from input video stream into
7345	separate grayscale video streams.
7346
7347	The filter accepts the following option:
7348
7349	@table @option
7350	@item planes
7351	Set plane(s) to extract.
7352
7353	Available values for planes are:
7354	@table @samp
7355	@item y
7356	@item u
7357	@item v
7358	@item a
7359	@item r
7360	@item g
7361	@item b
7362	@end table
7363
7364	Choosing planes not available in the input will result in an error.
7365	That means you cannot select @code{r}, @code{g}, @code{b} planes
7366	with @code{y}, @code{u}, @code{v} planes at same time.
7367	@end table
7368
7369	@subsection Examples
7370
7371	@itemize
7372	@item
7373	Extract luma, u and v color channel component from input video frame
7374	into 3 grayscale outputs:
7375	@example
7376	ffmpeg -i video.avi -filter_complex 'extractplanes=y+u+v[y][u][v]' -map '[y]' y.avi -map '[u]' u.avi -map '[v]' v.avi
7377	@end example
7378	@end itemize
7379
7380	@section elbg
7381
7382	Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
7383
7384	For each input image, the filter will compute the optimal mapping from
7385	the input to the output given the codebook length, that is the number
7386	of distinct output colors.
7387
7388	This filter accepts the following options.
7389
7390	@table @option
7391	@item codebook_length, l
7392	Set codebook length. The value must be a positive integer, and
7393	represents the number of distinct output colors. Default value is 256.
7394
7395	@item nb_steps, n
7396	Set the maximum number of iterations to apply for computing the optimal
7397	mapping. The higher the value the better the result and the higher the
7398	computation time. Default value is 1.
7399
7400	@item seed, s
7401	Set a random seed, must be an integer included between 0 and
7402	UINT32_MAX. If not specified, or if explicitly set to -1, the filter
7403	will try to use a good random seed on a best effort basis.
7404
7405	@item pal8
7406	Set pal8 output pixel format. This option does not work with codebook
7407	length greater than 256.
7408	@end table
7409
7410	@section fade
7411
7412	Apply a fade-in/out effect to the input video.
7413
7414	It accepts the following parameters:
7415
7416	@table @option
7417	@item type, t
7418	The effect type can be either "in" for a fade-in, or "out" for a fade-out
7419	effect.
7420	Default is @code{in}.
7421
7422	@item start_frame, s
7423	Specify the number of the frame to start applying the fade
7424	effect at. Default is 0.
7425
7426	@item nb_frames, n
7427	The number of frames that the fade effect lasts. At the end of the
7428	fade-in effect, the output video will have the same intensity as the input video.
7429	At the end of the fade-out transition, the output video will be filled with the
7430	selected @option{color}.
7431	Default is 25.
7432
7433	@item alpha
7434	If set to 1, fade only alpha channel, if one exists on the input.
7435	Default value is 0.
7436
7437	@item start_time, st
7438	Specify the timestamp (in seconds) of the frame to start to apply the fade
7439	effect. If both start_frame and start_time are specified, the fade will start at
7440	whichever comes last. Default is 0.
7441
7442	@item duration, d
7443	The number of seconds for which the fade effect has to last. At the end of the
7444	fade-in effect the output video will have the same intensity as the input video,
7445	at the end of the fade-out transition the output video will be filled with the
7446	selected @option{color}.
7447	If both duration and nb_frames are specified, duration is used. Default is 0
7448	(nb_frames is used by default).
7449
7450	@item color, c
7451	Specify the color of the fade. Default is "black".
7452	@end table
7453
7454	@subsection Examples
7455
7456	@itemize
7457	@item
7458	Fade in the first 30 frames of video:
7459	@example
7460	fade=in:0:30
7461	@end example
7462
7463	The command above is equivalent to:
7464	@example
7465	fade=t=in:s=0:n=30
7466	@end example
7467
7468	@item
7469	Fade out the last 45 frames of a 200-frame video:
7470	@example
7471	fade=out:155:45
7472	fade=type=out:start_frame=155:nb_frames=45
7473	@end example
7474
7475	@item
7476	Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
7477	@example
7478	fade=in:0:25, fade=out:975:25
7479	@end example
7480
7481	@item
7482	Make the first 5 frames yellow, then fade in from frame 5-24:
7483	@example
7484	fade=in:5:20:color=yellow
7485	@end example
7486
7487	@item
7488	Fade in alpha over first 25 frames of video:
7489	@example
7490	fade=in:0:25:alpha=1
7491	@end example
7492
7493	@item
7494	Make the first 5.5 seconds black, then fade in for 0.5 seconds:
7495	@example
7496	fade=t=in:st=5.5:d=0.5
7497	@end example
7498
7499	@end itemize
7500
7501	@section fftfilt
7502	Apply arbitrary expressions to samples in frequency domain
7503
7504	@table @option
7505	@item dc_Y
7506	Adjust the dc value (gain) of the luma plane of the image. The filter
7507	accepts an integer value in range @code{0} to @code{1000}. The default
7508	value is set to @code{0}.
7509
7510	@item dc_U
7511	Adjust the dc value (gain) of the 1st chroma plane of the image. The
7512	filter accepts an integer value in range @code{0} to @code{1000}. The
7513	default value is set to @code{0}.
7514
7515	@item dc_V
7516	Adjust the dc value (gain) of the 2nd chroma plane of the image. The
7517	filter accepts an integer value in range @code{0} to @code{1000}. The
7518	default value is set to @code{0}.
7519
7520	@item weight_Y
7521	Set the frequency domain weight expression for the luma plane.
7522
7523	@item weight_U
7524	Set the frequency domain weight expression for the 1st chroma plane.
7525
7526	@item weight_V
7527	Set the frequency domain weight expression for the 2nd chroma plane.
7528
7529	The filter accepts the following variables:
7530	@item X
7531	@item Y
7532	The coordinates of the current sample.
7533
7534	@item W
7535	@item H
7536	The width and height of the image.
7537	@end table
7538
7539	@subsection Examples
7540
7541	@itemize
7542	@item
7543	High-pass:
7544	@example
7545	fftfilt=dc_Y=128:weight_Y='squish(1-(Y+X)/100)'
7546	@end example
7547
7548	@item
7549	Low-pass:
7550	@example
7551	fftfilt=dc_Y=0:weight_Y='squish((Y+X)/100-1)'
7552	@end example
7553
7554	@item
7555	Sharpen:
7556	@example
7557	fftfilt=dc_Y=0:weight_Y='1+squish(1-(Y+X)/100)'
7558	@end example
7559
7560	@item
7561	Blur:
7562	@example
7563	fftfilt=dc_Y=0:weight_Y='exp(-4 * ((Y+X)/(W+H)))'
7564	@end example
7565
7566	@end itemize
7567
7568	@section field
7569
7570	Extract a single field from an interlaced image using stride
7571	arithmetic to avoid wasting CPU time. The output frames are marked as
7572	non-interlaced.
7573
7574	The filter accepts the following options:
7575
7576	@table @option
7577	@item type
7578	Specify whether to extract the top (if the value is @code{0} or
7579	@code{top}) or the bottom field (if the value is @code{1} or
7580	@code{bottom}).
7581	@end table
7582
7583	@section fieldhint
7584
7585	Create new frames by copying the top and bottom fields from surrounding frames
7586	supplied as numbers by the hint file.
7587
7588	@table @option
7589	@item hint
7590	Set file containing hints: absolute/relative frame numbers.
7591
7592	There must be one line for each frame in a clip. Each line must contain two
7593	numbers separated by the comma, optionally followed by @code{-} or @code{+}.
7594	Numbers supplied on each line of file can not be out of [N-1,N+1] where N
7595	is current frame number for @code{absolute} mode or out of [-1, 1] range
7596	for @code{relative} mode. First number tells from which frame to pick up top
7597	field and second number tells from which frame to pick up bottom field.
7598
7599	If optionally followed by @code{+} output frame will be marked as interlaced,
7600	else if followed by @code{-} output frame will be marked as progressive, else
7601	it will be marked same as input frame.
7602	If line starts with @code{#} or @code{;} that line is skipped.
7603
7604	@item mode
7605	Can be item @code{absolute} or @code{relative}. Default is @code{absolute}.
7606	@end table
7607
7608	Example of first several lines of @code{hint} file for @code{relative} mode:
7609	@example
7610	0,0 - # first frame
7611	1,0 - # second frame, use third's frame top field and second's frame bottom field
7612	1,0 - # third frame, use fourth's frame top field and third's frame bottom field
7613	1,0 -
7614	0,0 -
7615	0,0 -
7616	1,0 -
7617	1,0 -
7618	1,0 -
7619	0,0 -
7620	0,0 -
7621	1,0 -
7622	1,0 -
7623	1,0 -
7624	0,0 -
7625	@end example
7626
7627	@section fieldmatch
7628
7629	Field matching filter for inverse telecine. It is meant to reconstruct the
7630	progressive frames from a telecined stream. The filter does not drop duplicated
7631	frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
7632	followed by a decimation filter such as @ref{decimate} in the filtergraph.
7633
7634	The separation of the field matching and the decimation is notably motivated by
7635	the possibility of inserting a de-interlacing filter fallback between the two.
7636	If the source has mixed telecined and real interlaced content,
7637	@code{fieldmatch} will not be able to match fields for the interlaced parts.
7638	But these remaining combed frames will be marked as interlaced, and thus can be
7639	de-interlaced by a later filter such as @ref{yadif} before decimation.
7640
7641	In addition to the various configuration options, @code{fieldmatch} can take an
7642	optional second stream, activated through the @option{ppsrc} option. If
7643	enabled, the frames reconstruction will be based on the fields and frames from
7644	this second stream. This allows the first input to be pre-processed in order to
7645	help the various algorithms of the filter, while keeping the output lossless
7646	(assuming the fields are matched properly). Typically, a field-aware denoiser,
7647	or brightness/contrast adjustments can help.
7648
7649	Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
7650	and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
7651	which @code{fieldmatch} is based on. While the semantic and usage are very
7652	close, some behaviour and options names can differ.
7653
7654	The @ref{decimate} filter currently only works for constant frame rate input.
7655	If your input has mixed telecined (30fps) and progressive content with a lower
7656	framerate like 24fps use the following filterchain to produce the necessary cfr
7657	stream: @code{dejudder,fps=30000/1001,fieldmatch,decimate}.
7658
7659	The filter accepts the following options:
7660
7661	@table @option
7662	@item order
7663	Specify the assumed field order of the input stream. Available values are:
7664
7665	@table @samp
7666	@item auto
7667	Auto detect parity (use FFmpeg's internal parity value).
7668	@item bff
7669	Assume bottom field first.
7670	@item tff
7671	Assume top field first.
7672	@end table
7673
7674	Note that it is sometimes recommended not to trust the parity announced by the
7675	stream.
7676
7677	Default value is @var{auto}.
7678
7679	@item mode
7680	Set the matching mode or strategy to use. @option{pc} mode is the safest in the
7681	sense that it won't risk creating jerkiness due to duplicate frames when
7682	possible, but if there are bad edits or blended fields it will end up
7683	outputting combed frames when a good match might actually exist. On the other
7684	hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
7685	but will almost always find a good frame if there is one. The other values are
7686	all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
7687	jerkiness and creating duplicate frames versus finding good matches in sections
7688	with bad edits, orphaned fields, blended fields, etc.
7689
7690	More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
7691
7692	Available values are:
7693
7694	@table @samp
7695	@item pc
7696	2-way matching (p/c)
7697	@item pc_n
7698	2-way matching, and trying 3rd match if still combed (p/c + n)
7699	@item pc_u
7700	2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
7701	@item pc_n_ub
7702	2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
7703	still combed (p/c + n + u/b)
7704	@item pcn
7705	3-way matching (p/c/n)
7706	@item pcn_ub
7707	3-way matching, and trying 4th/5th matches if all 3 of the original matches are
7708	detected as combed (p/c/n + u/b)
7709	@end table
7710
7711	The parenthesis at the end indicate the matches that would be used for that
7712	mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
7713	@var{top}).
7714
7715	In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
7716	the slowest.
7717
7718	Default value is @var{pc_n}.
7719
7720	@item ppsrc
7721	Mark the main input stream as a pre-processed input, and enable the secondary
7722	input stream as the clean source to pick the fields from. See the filter
7723	introduction for more details. It is similar to the @option{clip2} feature from
7724	VFM/TFM.
7725
7726	Default value is @code{0} (disabled).
7727
7728	@item field
7729	Set the field to match from. It is recommended to set this to the same value as
7730	@option{order} unless you experience matching failures with that setting. In
7731	certain circumstances changing the field that is used to match from can have a
7732	large impact on matching performance. Available values are:
7733
7734	@table @samp
7735	@item auto
7736	Automatic (same value as @option{order}).
7737	@item bottom
7738	Match from the bottom field.
7739	@item top
7740	Match from the top field.
7741	@end table
7742
7743	Default value is @var{auto}.
7744
7745	@item mchroma
7746	Set whether or not chroma is included during the match comparisons. In most
7747	cases it is recommended to leave this enabled. You should set this to @code{0}
7748	only if your clip has bad chroma problems such as heavy rainbowing or other
7749	artifacts. Setting this to @code{0} could also be used to speed things up at
7750	the cost of some accuracy.
7751
7752	Default value is @code{1}.
7753
7754	@item y0
7755	@item y1
7756	These define an exclusion band which excludes the lines between @option{y0} and
7757	@option{y1} from being included in the field matching decision. An exclusion
7758	band can be used to ignore subtitles, a logo, or other things that may
7759	interfere with the matching. @option{y0} sets the starting scan line and
7760	@option{y1} sets the ending line; all lines in between @option{y0} and
7761	@option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
7762	@option{y0} and @option{y1} to the same value will disable the feature.
7763	@option{y0} and @option{y1} defaults to @code{0}.
7764
7765	@item scthresh
7766	Set the scene change detection threshold as a percentage of maximum change on
7767	the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
7768	detection is only relevant in case @option{combmatch}=@var{sc}. The range for
7769	@option{scthresh} is @code{[0.0, 100.0]}.
7770
7771	Default value is @code{12.0}.
7772
7773	@item combmatch
7774	When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
7775	account the combed scores of matches when deciding what match to use as the
7776	final match. Available values are:
7777
7778	@table @samp
7779	@item none
7780	No final matching based on combed scores.
7781	@item sc
7782	Combed scores are only used when a scene change is detected.
7783	@item full
7784	Use combed scores all the time.
7785	@end table
7786
7787	Default is @var{sc}.
7788
7789	@item combdbg
7790	Force @code{fieldmatch} to calculate the combed metrics for certain matches and
7791	print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
7792	Available values are:
7793
7794	@table @samp
7795	@item none
7796	No forced calculation.
7797	@item pcn
7798	Force p/c/n calculations.
7799	@item pcnub
7800	Force p/c/n/u/b calculations.
7801	@end table
7802
7803	Default value is @var{none}.
7804
7805	@item cthresh
7806	This is the area combing threshold used for combed frame detection. This
7807	essentially controls how "strong" or "visible" combing must be to be detected.
7808	Larger values mean combing must be more visible and smaller values mean combing
7809	can be less visible or strong and still be detected. Valid settings are from
7810	@code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
7811	be detected as combed). This is basically a pixel difference value. A good
7812	range is @code{[8, 12]}.
7813
7814	Default value is @code{9}.
7815
7816	@item chroma
7817	Sets whether or not chroma is considered in the combed frame decision. Only
7818	disable this if your source has chroma problems (rainbowing, etc.) that are
7819	causing problems for the combed frame detection with chroma enabled. Actually,
7820	using @option{chroma}=@var{0} is usually more reliable, except for the case
7821	where there is chroma only combing in the source.
7822
7823	Default value is @code{0}.
7824
7825	@item blockx
7826	@item blocky
7827	Respectively set the x-axis and y-axis size of the window used during combed
7828	frame detection. This has to do with the size of the area in which
7829	@option{combpel} pixels are required to be detected as combed for a frame to be
7830	declared combed. See the @option{combpel} parameter description for more info.
7831	Possible values are any number that is a power of 2 starting at 4 and going up
7832	to 512.
7833
7834	Default value is @code{16}.
7835
7836	@item combpel
7837	The number of combed pixels inside any of the @option{blocky} by
7838	@option{blockx} size blocks on the frame for the frame to be detected as
7839	combed. While @option{cthresh} controls how "visible" the combing must be, this
7840	setting controls "how much" combing there must be in any localized area (a
7841	window defined by the @option{blockx} and @option{blocky} settings) on the
7842	frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
7843	which point no frames will ever be detected as combed). This setting is known
7844	as @option{MI} in TFM/VFM vocabulary.
7845
7846	Default value is @code{80}.
7847	@end table
7848
7849	@anchor{p/c/n/u/b meaning}
7850	@subsection p/c/n/u/b meaning
7851
7852	@subsubsection p/c/n
7853
7854	We assume the following telecined stream:
7855
7856	@example
7857	Top fields: 1 2 2 3 4
7858	Bottom fields: 1 2 3 4 4
7859	@end example
7860
7861	The numbers correspond to the progressive frame the fields relate to. Here, the
7862	first two frames are progressive, the 3rd and 4th are combed, and so on.
7863
7864	When @code{fieldmatch} is configured to run a matching from bottom
7865	(@option{field}=@var{bottom}) this is how this input stream get transformed:
7866
7867	@example
7868	Input stream:
7869	T 1 2 2 3 4
7870	B 1 2 3 4 4 <-- matching reference
7871
7872	Matches: c c n n c
7873
7874	Output stream:
7875	T 1 2 3 4 4
7876	B 1 2 3 4 4
7877	@end example
7878
7879	As a result of the field matching, we can see that some frames get duplicated.
7880	To perform a complete inverse telecine, you need to rely on a decimation filter
7881	after this operation. See for instance the @ref{decimate} filter.
7882
7883	The same operation now matching from top fields (@option{field}=@var{top})
7884	looks like this:
7885
7886	@example
7887	Input stream:
7888	T 1 2 2 3 4 <-- matching reference
7889	B 1 2 3 4 4
7890
7891	Matches: c c p p c
7892
7893	Output stream:
7894	T 1 2 2 3 4
7895	B 1 2 2 3 4
7896	@end example
7897
7898	In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
7899	basically, they refer to the frame and field of the opposite parity:
7900
7901	@itemize
7902	@item @var{p} matches the field of the opposite parity in the previous frame
7903	@item @var{c} matches the field of the opposite parity in the current frame
7904	@item @var{n} matches the field of the opposite parity in the next frame
7905	@end itemize
7906
7907	@subsubsection u/b
7908
7909	The @var{u} and @var{b} matching are a bit special in the sense that they match
7910	from the opposite parity flag. In the following examples, we assume that we are
7911	currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
7912	'x' is placed above and below each matched fields.
7913
7914	With bottom matching (@option{field}=@var{bottom}):
7915	@example
7916	Match: c p n b u
7917
7918	x x x x x
7919	Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7920	Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
7921	x x x x x
7922
7923	Output frames:
7924	2 1 2 2 2
7925	2 2 2 1 3
7926	@end example
7927
7928	With top matching (@option{field}=@var{top}):
7929	@example
7930	Match: c p n b u
7931
7932	x x x x x
7933	Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7934	Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
7935	x x x x x
7936
7937	Output frames:
7938	2 2 2 1 2
7939	2 1 3 2 2
7940	@end example
7941
7942	@subsection Examples
7943
7944	Simple IVTC of a top field first telecined stream:
7945	@example
7946	fieldmatch=order=tff:combmatch=none, decimate
7947	@end example
7948
7949	Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
7950	@example
7951	fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
7952	@end example
7953
7954	@section fieldorder
7955
7956	Transform the field order of the input video.
7957
7958	It accepts the following parameters:
7959
7960	@table @option
7961
7962	@item order
7963	The output field order. Valid values are @var{tff} for top field first or @var{bff}
7964	for bottom field first.
7965	@end table
7966
7967	The default value is @samp{tff}.
7968
7969	The transformation is done by shifting the picture content up or down
7970	by one line, and filling the remaining line with appropriate picture content.
7971	This method is consistent with most broadcast field order converters.
7972
7973	If the input video is not flagged as being interlaced, or it is already
7974	flagged as being of the required output field order, then this filter does
7975	not alter the incoming video.
7976
7977	It is very useful when converting to or from PAL DV material,
7978	which is bottom field first.
7979
7980	For example:
7981	@example
7982	ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
7983	@end example
7984
7985	@section fifo, afifo
7986
7987	Buffer input images and send them when they are requested.
7988
7989	It is mainly useful when auto-inserted by the libavfilter
7990	framework.
7991
7992	It does not take parameters.
7993
7994	@section find_rect
7995
7996	Find a rectangular object
7997
7998	It accepts the following options:
7999
8000	@table @option
8001	@item object
8002	Filepath of the object image, needs to be in gray8.
8003
8004	@item threshold
8005	Detection threshold, default is 0.5.
8006
8007	@item mipmaps
8008	Number of mipmaps, default is 3.
8009
8010	@item xmin, ymin, xmax, ymax
8011	Specifies the rectangle in which to search.
8012	@end table
8013
8014	@subsection Examples
8015
8016	@itemize
8017	@item
8018	Generate a representative palette of a given video using @command{ffmpeg}:
8019	@example
8020	ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
8021	@end example
8022	@end itemize
8023
8024	@section cover_rect
8025
8026	Cover a rectangular object
8027
8028	It accepts the following options:
8029
8030	@table @option
8031	@item cover
8032	Filepath of the optional cover image, needs to be in yuv420.
8033
8034	@item mode
8035	Set covering mode.
8036
8037	It accepts the following values:
8038	@table @samp
8039	@item cover
8040	cover it by the supplied image
8041	@item blur
8042	cover it by interpolating the surrounding pixels
8043	@end table
8044
8045	Default value is @var{blur}.
8046	@end table
8047
8048	@subsection Examples
8049
8050	@itemize
8051	@item
8052	Generate a representative palette of a given video using @command{ffmpeg}:
8053	@example
8054	ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
8055	@end example
8056	@end itemize
8057
8058	@anchor{format}
8059	@section format
8060
8061	Convert the input video to one of the specified pixel formats.
8062	Libavfilter will try to pick one that is suitable as input to
8063	the next filter.
8064
8065	It accepts the following parameters:
8066	@table @option
8067
8068	@item pix_fmts
8069	A '|'-separated list of pixel format names, such as
8070	"pix_fmts=yuv420p|monow|rgb24".
8071
8072	@end table
8073
8074	@subsection Examples
8075
8076	@itemize
8077	@item
8078	Convert the input video to the @var{yuv420p} format
8079	@example
8080	format=pix_fmts=yuv420p
8081	@end example
8082
8083	Convert the input video to any of the formats in the list
8084	@example
8085	format=pix_fmts=yuv420p|yuv444p|yuv410p
8086	@end example
8087	@end itemize
8088
8089	@anchor{fps}
8090	@section fps
8091
8092	Convert the video to specified constant frame rate by duplicating or dropping
8093	frames as necessary.
8094
8095	It accepts the following parameters:
8096	@table @option
8097
8098	@item fps
8099	The desired output frame rate. The default is @code{25}.
8100
8101	@item round
8102	Rounding method.
8103
8104	Possible values are:
8105	@table @option
8106	@item zero
8107	zero round towards 0
8108	@item inf
8109	round away from 0
8110	@item down
8111	round towards -infinity
8112	@item up
8113	round towards +infinity
8114	@item near
8115	round to nearest
8116	@end table
8117	The default is @code{near}.
8118
8119	@item start_time
8120	Assume the first PTS should be the given value, in seconds. This allows for
8121	padding/trimming at the start of stream. By default, no assumption is made
8122	about the first frame's expected PTS, so no padding or trimming is done.
8123	For example, this could be set to 0 to pad the beginning with duplicates of
8124	the first frame if a video stream starts after the audio stream or to trim any
8125	frames with a negative PTS.
8126
8127	@end table
8128
8129	Alternatively, the options can be specified as a flat string:
8130	@var{fps}[:@var{round}].
8131
8132	See also the @ref{setpts} filter.
8133
8134	@subsection Examples
8135
8136	@itemize
8137	@item
8138	A typical usage in order to set the fps to 25:
8139	@example
8140	fps=fps=25
8141	@end example
8142
8143	@item
8144	Sets the fps to 24, using abbreviation and rounding method to round to nearest:
8145	@example
8146	fps=fps=film:round=near
8147	@end example
8148	@end itemize
8149
8150	@section framepack
8151
8152	Pack two different video streams into a stereoscopic video, setting proper
8153	metadata on supported codecs. The two views should have the same size and
8154	framerate and processing will stop when the shorter video ends. Please note
8155	that you may conveniently adjust view properties with the @ref{scale} and
8156	@ref{fps} filters.
8157
8158	It accepts the following parameters:
8159	@table @option
8160
8161	@item format
8162	The desired packing format. Supported values are:
8163
8164	@table @option
8165
8166	@item sbs
8167	The views are next to each other (default).
8168
8169	@item tab
8170	The views are on top of each other.
8171
8172	@item lines
8173	The views are packed by line.
8174
8175	@item columns
8176	The views are packed by column.
8177
8178	@item frameseq
8179	The views are temporally interleaved.
8180
8181	@end table
8182
8183	@end table
8184
8185	Some examples:
8186
8187	@example
8188	# Convert left and right views into a frame-sequential video
8189	ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
8190
8191	# Convert views into a side-by-side video with the same output resolution as the input
8192	ffmpeg -i LEFT -i RIGHT -filter_complex [0:v]scale=w=iw/2[left],[1:v]scale=w=iw/2[right],[left][right]framepack=sbs OUTPUT
8193	@end example
8194
8195	@section framerate
8196
8197	Change the frame rate by interpolating new video output frames from the source
8198	frames.
8199
8200	This filter is not designed to function correctly with interlaced media. If
8201	you wish to change the frame rate of interlaced media then you are required
8202	to deinterlace before this filter and re-interlace after this filter.
8203
8204	A description of the accepted options follows.
8205
8206	@table @option
8207	@item fps
8208	Specify the output frames per second. This option can also be specified
8209	as a value alone. The default is @code{50}.
8210
8211	@item interp_start
8212	Specify the start of a range where the output frame will be created as a
8213	linear interpolation of two frames. The range is [@code{0}-@code{255}],
8214	the default is @code{15}.
8215
8216	@item interp_end
8217	Specify the end of a range where the output frame will be created as a
8218	linear interpolation of two frames. The range is [@code{0}-@code{255}],
8219	the default is @code{240}.
8220
8221	@item scene
8222	Specify the level at which a scene change is detected as a value between
8223	0 and 100 to indicate a new scene; a low value reflects a low
8224	probability for the current frame to introduce a new scene, while a higher
8225	value means the current frame is more likely to be one.
8226	The default is @code{7}.
8227
8228	@item flags
8229	Specify flags influencing the filter process.
8230
8231	Available value for @var{flags} is:
8232
8233	@table @option
8234	@item scene_change_detect, scd
8235	Enable scene change detection using the value of the option @var{scene}.
8236	This flag is enabled by default.
8237	@end table
8238	@end table
8239
8240	@section framestep
8241
8242	Select one frame every N-th frame.
8243
8244	This filter accepts the following option:
8245	@table @option
8246	@item step
8247	Select frame after every @code{step} frames.
8248	Allowed values are positive integers higher than 0. Default value is @code{1}.
8249	@end table
8250
8251	@anchor{frei0r}
8252	@section frei0r
8253
8254	Apply a frei0r effect to the input video.
8255
8256	To enable the compilation of this filter, you need to install the frei0r
8257	header and configure FFmpeg with @code{--enable-frei0r}.
8258
8259	It accepts the following parameters:
8260
8261	@table @option
8262
8263	@item filter_name
8264	The name of the frei0r effect to load. If the environment variable
8265	@env{FREI0R_PATH} is defined, the frei0r effect is searched for in each of the
8266	directories specified by the colon-separated list in @env{FREIOR_PATH}.
8267	Otherwise, the standard frei0r paths are searched, in this order:
8268	@file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
8269	@file{/usr/lib/frei0r-1/}.
8270
8271	@item filter_params
8272	A '|'-separated list of parameters to pass to the frei0r effect.
8273
8274	@end table
8275
8276	A frei0r effect parameter can be a boolean (its value is either
8277	"y" or "n"), a double, a color (specified as
8278	@var{R}/@var{G}/@var{B}, where @var{R}, @var{G}, and @var{B} are floating point
8279	numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
8280	section in the ffmpeg-utils manual), a position (specified as @var{X}/@var{Y}, where
8281	@var{X} and @var{Y} are floating point numbers) and/or a string.
8282
8283	The number and types of parameters depend on the loaded effect. If an
8284	effect parameter is not specified, the default value is set.
8285
8286	@subsection Examples
8287
8288	@itemize
8289	@item
8290	Apply the distort0r effect, setting the first two double parameters:
8291	@example
8292	frei0r=filter_name=distort0r:filter_params=0.5|0.01
8293	@end example
8294
8295	@item
8296	Apply the colordistance effect, taking a color as the first parameter:
8297	@example
8298	frei0r=colordistance:0.2/0.3/0.4
8299	frei0r=colordistance:violet
8300	frei0r=colordistance:0x112233
8301	@end example
8302
8303	@item
8304	Apply the perspective effect, specifying the top left and top right image
8305	positions:
8306	@example
8307	frei0r=perspective:0.2/0.2|0.8/0.2
8308	@end example
8309	@end itemize
8310
8311	For more information, see
8312	@url{http://frei0r.dyne.org}
8313
8314	@section fspp
8315
8316	Apply fast and simple postprocessing. It is a faster version of @ref{spp}.
8317
8318	It splits (I)DCT into horizontal/vertical passes. Unlike the simple post-
8319	processing filter, one of them is performed once per block, not per pixel.
8320	This allows for much higher speed.
8321
8322	The filter accepts the following options:
8323
8324	@table @option
8325	@item quality
8326	Set quality. This option defines the number of levels for averaging. It accepts
8327	an integer in the range 4-5. Default value is @code{4}.
8328
8329	@item qp
8330	Force a constant quantization parameter. It accepts an integer in range 0-63.
8331	If not set, the filter will use the QP from the video stream (if available).
8332
8333	@item strength
8334	Set filter strength. It accepts an integer in range -15 to 32. Lower values mean
8335	more details but also more artifacts, while higher values make the image smoother
8336	but also blurrier. Default value is @code{0} − PSNR optimal.
8337
8338	@item use_bframe_qp
8339	Enable the use of the QP from the B-Frames if set to @code{1}. Using this
8340	option may cause flicker since the B-Frames have often larger QP. Default is
8341	@code{0} (not enabled).
8342
8343	@end table
8344
8345	@section gblur
8346
8347	Apply Gaussian blur filter.
8348
8349	The filter accepts the following options:
8350
8351	@table @option
8352	@item sigma
8353	Set horizontal sigma, standard deviation of Gaussian blur. Default is @code{0.5}.
8354
8355	@item steps
8356	Set number of steps for Gaussian approximation. Defauls is @code{1}.
8357
8358	@item planes
8359	Set which planes to filter. By default all planes are filtered.
8360
8361	@item sigmaV
8362	Set vertical sigma, if negative it will be same as @code{sigma}.
8363	Default is @code{-1}.
8364	@end table
8365
8366	@section geq
8367
8368	The filter accepts the following options:
8369
8370	@table @option
8371	@item lum_expr, lum
8372	Set the luminance expression.
8373	@item cb_expr, cb
8374	Set the chrominance blue expression.
8375	@item cr_expr, cr
8376	Set the chrominance red expression.
8377	@item alpha_expr, a
8378	Set the alpha expression.
8379	@item red_expr, r
8380	Set the red expression.
8381	@item green_expr, g
8382	Set the green expression.
8383	@item blue_expr, b
8384	Set the blue expression.
8385	@end table
8386
8387	The colorspace is selected according to the specified options. If one
8388	of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
8389	options is specified, the filter will automatically select a YCbCr
8390	colorspace. If one of the @option{red_expr}, @option{green_expr}, or
8391	@option{blue_expr} options is specified, it will select an RGB
8392	colorspace.
8393
8394	If one of the chrominance expression is not defined, it falls back on the other
8395	one. If no alpha expression is specified it will evaluate to opaque value.
8396	If none of chrominance expressions are specified, they will evaluate
8397	to the luminance expression.
8398
8399	The expressions can use the following variables and functions:
8400
8401	@table @option
8402	@item N
8403	The sequential number of the filtered frame, starting from @code{0}.
8404
8405	@item X
8406	@item Y
8407	The coordinates of the current sample.
8408
8409	@item W
8410	@item H
8411	The width and height of the image.
8412
8413	@item SW
8414	@item SH
8415	Width and height scale depending on the currently filtered plane. It is the
8416	ratio between the corresponding luma plane number of pixels and the current
8417	plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
8418	@code{0.5,0.5} for chroma planes.
8419
8420	@item T
8421	Time of the current frame, expressed in seconds.
8422
8423	@item p(x, y)
8424	Return the value of the pixel at location (@var{x},@var{y}) of the current
8425	plane.
8426
8427	@item lum(x, y)
8428	Return the value of the pixel at location (@var{x},@var{y}) of the luminance
8429	plane.
8430
8431	@item cb(x, y)
8432	Return the value of the pixel at location (@var{x},@var{y}) of the
8433	blue-difference chroma plane. Return 0 if there is no such plane.
8434
8435	@item cr(x, y)
8436	Return the value of the pixel at location (@var{x},@var{y}) of the
8437	red-difference chroma plane. Return 0 if there is no such plane.
8438
8439	@item r(x, y)
8440	@item g(x, y)
8441	@item b(x, y)
8442	Return the value of the pixel at location (@var{x},@var{y}) of the
8443	red/green/blue component. Return 0 if there is no such component.
8444
8445	@item alpha(x, y)
8446	Return the value of the pixel at location (@var{x},@var{y}) of the alpha
8447	plane. Return 0 if there is no such plane.
8448	@end table
8449
8450	For functions, if @var{x} and @var{y} are outside the area, the value will be
8451	automatically clipped to the closer edge.
8452
8453	@subsection Examples
8454
8455	@itemize
8456	@item
8457	Flip the image horizontally:
8458	@example
8459	geq=p(W-X\,Y)
8460	@end example
8461
8462	@item
8463	Generate a bidimensional sine wave, with angle @code{PI/3} and a
8464	wavelength of 100 pixels:
8465	@example
8466	geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
8467	@end example
8468
8469	@item
8470	Generate a fancy enigmatic moving light:
8471	@example
8472	nullsrc=s=256x256,geq=random(1)/hypot(X-cos(N*0.07)*W/2-W/2\,Y-sin(N*0.09)*H/2-H/2)^2*1000000*sin(N*0.02):128:128
8473	@end example
8474
8475	@item
8476	Generate a quick emboss effect:
8477	@example
8478	format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
8479	@end example
8480
8481	@item
8482	Modify RGB components depending on pixel position:
8483	@example
8484	geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
8485	@end example
8486
8487	@item
8488	Create a radial gradient that is the same size as the input (also see
8489	the @ref{vignette} filter):
8490	@example
8491	geq=lum=255*gauss((X/W-0.5)*3)*gauss((Y/H-0.5)*3)/gauss(0)/gauss(0),format=gray
8492	@end example
8493	@end itemize
8494
8495	@section gradfun
8496
8497	Fix the banding artifacts that are sometimes introduced into nearly flat
8498	regions by truncation to 8-bit color depth.
8499	Interpolate the gradients that should go where the bands are, and
8500	dither them.
8501
8502	It is designed for playback only. Do not use it prior to
8503	lossy compression, because compression tends to lose the dither and
8504	bring back the bands.
8505
8506	It accepts the following parameters:
8507
8508	@table @option
8509
8510	@item strength
8511	The maximum amount by which the filter will change any one pixel. This is also
8512	the threshold for detecting nearly flat regions. Acceptable values range from
8513	.51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
8514	valid range.
8515
8516	@item radius
8517	The neighborhood to fit the gradient to. A larger radius makes for smoother
8518	gradients, but also prevents the filter from modifying the pixels near detailed
8519	regions. Acceptable values are 8-32; the default value is 16. Out-of-range
8520	values will be clipped to the valid range.
8521
8522	@end table
8523
8524	Alternatively, the options can be specified as a flat string:
8525	@var{strength}[:@var{radius}]
8526
8527	@subsection Examples
8528
8529	@itemize
8530	@item
8531	Apply the filter with a @code{3.5} strength and radius of @code{8}:
8532	@example
8533	gradfun=3.5:8
8534	@end example
8535
8536	@item
8537	Specify radius, omitting the strength (which will fall-back to the default
8538	value):
8539	@example
8540	gradfun=radius=8
8541	@end example
8542
8543	@end itemize
8544
8545	@anchor{haldclut}
8546	@section haldclut
8547
8548	Apply a Hald CLUT to a video stream.
8549
8550	First input is the video stream to process, and second one is the Hald CLUT.
8551	The Hald CLUT input can be a simple picture or a complete video stream.
8552
8553	The filter accepts the following options:
8554
8555	@table @option
8556	@item shortest
8557	Force termination when the shortest input terminates. Default is @code{0}.
8558	@item repeatlast
8559	Continue applying the last CLUT after the end of the stream. A value of
8560	@code{0} disable the filter after the last frame of the CLUT is reached.
8561	Default is @code{1}.
8562	@end table
8563
8564	@code{haldclut} also has the same interpolation options as @ref{lut3d} (both
8565	filters share the same internals).
8566
8567	More information about the Hald CLUT can be found on Eskil Steenberg's website
8568	(Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
8569
8570	@subsection Workflow examples
8571
8572	@subsubsection Hald CLUT video stream
8573
8574	Generate an identity Hald CLUT stream altered with various effects:
8575	@example
8576	ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "hue=H=2*PI*t:s=sin(2*PI*t)+1, curves=cross_process" -t 10 -c:v ffv1 clut.nut
8577	@end example
8578
8579	Note: make sure you use a lossless codec.
8580
8581	Then use it with @code{haldclut} to apply it on some random stream:
8582	@example
8583	ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
8584	@end example
8585
8586	The Hald CLUT will be applied to the 10 first seconds (duration of
8587	@file{clut.nut}), then the latest picture of that CLUT stream will be applied
8588	to the remaining frames of the @code{mandelbrot} stream.
8589
8590	@subsubsection Hald CLUT with preview
8591
8592	A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
8593	@code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
8594	biggest possible square starting at the top left of the picture. The remaining
8595	padding pixels (bottom or right) will be ignored. This area can be used to add
8596	a preview of the Hald CLUT.
8597
8598	Typically, the following generated Hald CLUT will be supported by the
8599	@code{haldclut} filter:
8600
8601	@example
8602	ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
8603	pad=iw+320 [padded_clut];
8604	smptebars=s=320x256, split [a][b];
8605	[padded_clut][a] overlay=W-320:h, curves=color_negative [main];
8606	[main][b] overlay=W-320" -frames:v 1 clut.png
8607	@end example
8608
8609	It contains the original and a preview of the effect of the CLUT: SMPTE color
8610	bars are displayed on the right-top, and below the same color bars processed by
8611	the color changes.
8612
8613	Then, the effect of this Hald CLUT can be visualized with:
8614	@example
8615	ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
8616	@end example
8617
8618	@section hflip
8619
8620	Flip the input video horizontally.
8621
8622	For example, to horizontally flip the input video with @command{ffmpeg}:
8623	@example
8624	ffmpeg -i in.avi -vf "hflip" out.avi
8625	@end example
8626
8627	@section histeq
8628	This filter applies a global color histogram equalization on a
8629	per-frame basis.
8630
8631	It can be used to correct video that has a compressed range of pixel
8632	intensities. The filter redistributes the pixel intensities to
8633	equalize their distribution across the intensity range. It may be
8634	viewed as an "automatically adjusting contrast filter". This filter is
8635	useful only for correcting degraded or poorly captured source
8636	video.
8637
8638	The filter accepts the following options:
8639
8640	@table @option
8641	@item strength
8642	Determine the amount of equalization to be applied. As the strength
8643	is reduced, the distribution of pixel intensities more-and-more
8644	approaches that of the input frame. The value must be a float number
8645	in the range [0,1] and defaults to 0.200.
8646
8647	@item intensity
8648	Set the maximum intensity that can generated and scale the output
8649	values appropriately. The strength should be set as desired and then
8650	the intensity can be limited if needed to avoid washing-out. The value
8651	must be a float number in the range [0,1] and defaults to 0.210.
8652
8653	@item antibanding
8654	Set the antibanding level. If enabled the filter will randomly vary
8655	the luminance of output pixels by a small amount to avoid banding of
8656	the histogram. Possible values are @code{none}, @code{weak} or
8657	@code{strong}. It defaults to @code{none}.
8658	@end table
8659
8660	@section histogram
8661
8662	Compute and draw a color distribution histogram for the input video.
8663
8664	The computed histogram is a representation of the color component
8665	distribution in an image.
8666
8667	Standard histogram displays the color components distribution in an image.
8668	Displays color graph for each color component. Shows distribution of
8669	the Y, U, V, A or R, G, B components, depending on input format, in the
8670	current frame. Below each graph a color component scale meter is shown.
8671
8672	The filter accepts the following options:
8673
8674	@table @option
8675	@item level_height
8676	Set height of level. Default value is @code{200}.
8677	Allowed range is [50, 2048].
8678
8679	@item scale_height
8680	Set height of color scale. Default value is @code{12}.
8681	Allowed range is [0, 40].
8682
8683	@item display_mode
8684	Set display mode.
8685	It accepts the following values:
8686	@table @samp
8687	@item parade
8688	Per color component graphs are placed below each other.
8689
8690	@item overlay
8691	Presents information identical to that in the @code{parade}, except
8692	that the graphs representing color components are superimposed directly
8693	over one another.
8694	@end table
8695	Default is @code{parade}.
8696
8697	@item levels_mode
8698	Set mode. Can be either @code{linear}, or @code{logarithmic}.
8699	Default is @code{linear}.
8700
8701	@item components
8702	Set what color components to display.
8703	Default is @code{7}.
8704
8705	@item fgopacity
8706	Set foreground opacity. Default is @code{0.7}.
8707
8708	@item bgopacity
8709	Set background opacity. Default is @code{0.5}.
8710	@end table
8711
8712	@subsection Examples
8713
8714	@itemize
8715
8716	@item
8717	Calculate and draw histogram:
8718	@example
8719	ffplay -i input -vf histogram
8720	@end example
8721
8722	@end itemize
8723
8724	@anchor{hqdn3d}
8725	@section hqdn3d
8726
8727	This is a high precision/quality 3d denoise filter. It aims to reduce
8728	image noise, producing smooth images and making still images really
8729	still. It should enhance compressibility.
8730
8731	It accepts the following optional parameters:
8732
8733	@table @option
8734	@item luma_spatial
8735	A non-negative floating point number which specifies spatial luma strength.
8736	It defaults to 4.0.
8737
8738	@item chroma_spatial
8739	A non-negative floating point number which specifies spatial chroma strength.
8740	It defaults to 3.0*@var{luma_spatial}/4.0.
8741
8742	@item luma_tmp
8743	A floating point number which specifies luma temporal strength. It defaults to
8744	6.0*@var{luma_spatial}/4.0.
8745
8746	@item chroma_tmp
8747	A floating point number which specifies chroma temporal strength. It defaults to
8748	@var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}.
8749	@end table
8750
8751	@anchor{hwupload_cuda}
8752	@section hwupload_cuda
8753
8754	Upload system memory frames to a CUDA device.
8755
8756	It accepts the following optional parameters:
8757
8758	@table @option
8759	@item device
8760	The number of the CUDA device to use
8761	@end table
8762
8763	@section hqx
8764
8765	Apply a high-quality magnification filter designed for pixel art. This filter
8766	was originally created by Maxim Stepin.
8767
8768	It accepts the following option:
8769
8770	@table @option
8771	@item n
8772	Set the scaling dimension: @code{2} for @code{hq2x}, @code{3} for
8773	@code{hq3x} and @code{4} for @code{hq4x}.
8774	Default is @code{3}.
8775	@end table
8776
8777	@section hstack
8778	Stack input videos horizontally.
8779
8780	All streams must be of same pixel format and of same height.
8781
8782	Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
8783	to create same output.
8784
8785	The filter accept the following option:
8786
8787	@table @option
8788	@item inputs
8789	Set number of input streams. Default is 2.
8790
8791	@item shortest
8792	If set to 1, force the output to terminate when the shortest input
8793	terminates. Default value is 0.
8794	@end table
8795
8796	@section hue
8797
8798	Modify the hue and/or the saturation of the input.
8799
8800	It accepts the following parameters:
8801
8802	@table @option
8803	@item h
8804	Specify the hue angle as a number of degrees. It accepts an expression,
8805	and defaults to "0".
8806
8807	@item s
8808	Specify the saturation in the [-10,10] range. It accepts an expression and
8809	defaults to "1".
8810
8811	@item H
8812	Specify the hue angle as a number of radians. It accepts an
8813	expression, and defaults to "0".
8814
8815	@item b
8816	Specify the brightness in the [-10,10] range. It accepts an expression and
8817	defaults to "0".
8818	@end table
8819
8820	@option{h} and @option{H} are mutually exclusive, and can't be
8821	specified at the same time.
8822
8823	The @option{b}, @option{h}, @option{H} and @option{s} option values are
8824	expressions containing the following constants:
8825
8826	@table @option
8827	@item n
8828	frame count of the input frame starting from 0
8829
8830	@item pts
8831	presentation timestamp of the input frame expressed in time base units
8832
8833	@item r
8834	frame rate of the input video, NAN if the input frame rate is unknown
8835
8836	@item t
8837	timestamp expressed in seconds, NAN if the input timestamp is unknown
8838
8839	@item tb
8840	time base of the input video
8841	@end table
8842
8843	@subsection Examples
8844
8845	@itemize
8846	@item
8847	Set the hue to 90 degrees and the saturation to 1.0:
8848	@example
8849	hue=h=90:s=1
8850	@end example
8851
8852	@item
8853	Same command but expressing the hue in radians:
8854	@example
8855	hue=H=PI/2:s=1
8856	@end example
8857
8858	@item
8859	Rotate hue and make the saturation swing between 0
8860	and 2 over a period of 1 second:
8861	@example
8862	hue="H=2*PI*t: s=sin(2*PI*t)+1"
8863	@end example
8864
8865	@item
8866	Apply a 3 seconds saturation fade-in effect starting at 0:
8867	@example
8868	hue="s=min(t/3\,1)"
8869	@end example
8870
8871	The general fade-in expression can be written as:
8872	@example
8873	hue="s=min(0\, max((t-START)/DURATION\, 1))"
8874	@end example
8875
8876	@item
8877	Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
8878	@example
8879	hue="s=max(0\, min(1\, (8-t)/3))"
8880	@end example
8881
8882	The general fade-out expression can be written as:
8883	@example
8884	hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
8885	@end example
8886
8887	@end itemize
8888
8889	@subsection Commands
8890
8891	This filter supports the following commands:
8892	@table @option
8893	@item b
8894	@item s
8895	@item h
8896	@item H
8897	Modify the hue and/or the saturation and/or brightness of the input video.
8898	The command accepts the same syntax of the corresponding option.
8899
8900	If the specified expression is not valid, it is kept at its current
8901	value.
8902	@end table
8903
8904	@section hysteresis
8905
8906	Grow first stream into second stream by connecting components.
8907	This makes it possible to build more robust edge masks.
8908
8909	This filter accepts the following options:
8910
8911	@table @option
8912	@item planes
8913	Set which planes will be processed as bitmap, unprocessed planes will be
8914	copied from first stream.
8915	By default value 0xf, all planes will be processed.
8916
8917	@item threshold
8918	Set threshold which is used in filtering. If pixel component value is higher than
8919	this value filter algorithm for connecting components is activated.
8920	By default value is 0.
8921	@end table
8922
8923	@section idet
8924
8925	Detect video interlacing type.
8926
8927	This filter tries to detect if the input frames are interlaced, progressive,
8928	top or bottom field first. It will also try to detect fields that are
8929	repeated between adjacent frames (a sign of telecine).
8930
8931	Single frame detection considers only immediately adjacent frames when classifying each frame.
8932	Multiple frame detection incorporates the classification history of previous frames.
8933
8934	The filter will log these metadata values:
8935
8936	@table @option
8937	@item single.current_frame
8938	Detected type of current frame using single-frame detection. One of:
8939	``tff'' (top field first), ``bff'' (bottom field first),
8940	``progressive'', or ``undetermined''
8941
8942	@item single.tff
8943	Cumulative number of frames detected as top field first using single-frame detection.
8944
8945	@item multiple.tff
8946	Cumulative number of frames detected as top field first using multiple-frame detection.
8947
8948	@item single.bff
8949	Cumulative number of frames detected as bottom field first using single-frame detection.
8950
8951	@item multiple.current_frame
8952	Detected type of current frame using multiple-frame detection. One of:
8953	``tff'' (top field first), ``bff'' (bottom field first),
8954	``progressive'', or ``undetermined''
8955
8956	@item multiple.bff
8957	Cumulative number of frames detected as bottom field first using multiple-frame detection.
8958
8959	@item single.progressive
8960	Cumulative number of frames detected as progressive using single-frame detection.
8961
8962	@item multiple.progressive
8963	Cumulative number of frames detected as progressive using multiple-frame detection.
8964
8965	@item single.undetermined
8966	Cumulative number of frames that could not be classified using single-frame detection.
8967
8968	@item multiple.undetermined
8969	Cumulative number of frames that could not be classified using multiple-frame detection.
8970
8971	@item repeated.current_frame
8972	Which field in the current frame is repeated from the last. One of ``neither'', ``top'', or ``bottom''.
8973
8974	@item repeated.neither
8975	Cumulative number of frames with no repeated field.
8976
8977	@item repeated.top
8978	Cumulative number of frames with the top field repeated from the previous frame's top field.
8979
8980	@item repeated.bottom
8981	Cumulative number of frames with the bottom field repeated from the previous frame's bottom field.
8982	@end table
8983
8984	The filter accepts the following options:
8985
8986	@table @option
8987	@item intl_thres
8988	Set interlacing threshold.
8989	@item prog_thres
8990	Set progressive threshold.
8991	@item rep_thres
8992	Threshold for repeated field detection.
8993	@item half_life
8994	Number of frames after which a given frame's contribution to the
8995	statistics is halved (i.e., it contributes only 0.5 to its
8996	classification). The default of 0 means that all frames seen are given
8997	full weight of 1.0 forever.
8998	@item analyze_interlaced_flag
8999	When this is not 0 then idet will use the specified number of frames to determine
9000	if the interlaced flag is accurate, it will not count undetermined frames.
9001	If the flag is found to be accurate it will be used without any further
9002	computations, if it is found to be inaccurate it will be cleared without any
9003	further computations. This allows inserting the idet filter as a low computational
9004	method to clean up the interlaced flag
9005	@end table
9006
9007	@section il
9008
9009	Deinterleave or interleave fields.
9010
9011	This filter allows one to process interlaced images fields without
9012	deinterlacing them. Deinterleaving splits the input frame into 2
9013	fields (so called half pictures). Odd lines are moved to the top
9014	half of the output image, even lines to the bottom half.
9015	You can process (filter) them independently and then re-interleave them.
9016
9017	The filter accepts the following options:
9018
9019	@table @option
9020	@item luma_mode, l
9021	@item chroma_mode, c
9022	@item alpha_mode, a
9023	Available values for @var{luma_mode}, @var{chroma_mode} and
9024	@var{alpha_mode} are:
9025
9026	@table @samp
9027	@item none
9028	Do nothing.
9029
9030	@item deinterleave, d
9031	Deinterleave fields, placing one above the other.
9032
9033	@item interleave, i
9034	Interleave fields. Reverse the effect of deinterleaving.
9035	@end table
9036	Default value is @code{none}.
9037
9038	@item luma_swap, ls
9039	@item chroma_swap, cs
9040	@item alpha_swap, as
9041	Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
9042	@end table
9043
9044	@section inflate
9045
9046	Apply inflate effect to the video.
9047
9048	This filter replaces the pixel by the local(3x3) average by taking into account
9049	only values higher than the pixel.
9050
9051	It accepts the following options:
9052
9053	@table @option
9054	@item threshold0
9055	@item threshold1
9056	@item threshold2
9057	@item threshold3
9058	Limit the maximum change for each plane, default is 65535.
9059	If 0, plane will remain unchanged.
9060	@end table
9061
9062	@section interlace
9063
9064	Simple interlacing filter from progressive contents. This interleaves upper (or
9065	lower) lines from odd frames with lower (or upper) lines from even frames,
9066	halving the frame rate and preserving image height.
9067
9068	@example
9069	Original Original New Frame
9070	Frame 'j' Frame 'j+1' (tff)
9071	========== =========== ==================
9072	Line 0 --------------------> Frame 'j' Line 0
9073	Line 1 Line 1 ----> Frame 'j+1' Line 1
9074	Line 2 ---------------------> Frame 'j' Line 2
9075	Line 3 Line 3 ----> Frame 'j+1' Line 3
9076	... ... ...
9077	New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
9078	@end example
9079
9080	It accepts the following optional parameters:
9081
9082	@table @option
9083	@item scan
9084	This determines whether the interlaced frame is taken from the even
9085	(tff - default) or odd (bff) lines of the progressive frame.
9086
9087	@item lowpass
9088	Enable (default) or disable the vertical lowpass filter to avoid twitter
9089	interlacing and reduce moire patterns.
9090	@end table
9091
9092	@section kerndeint
9093
9094	Deinterlace input video by applying Donald Graft's adaptive kernel
9095	deinterling. Work on interlaced parts of a video to produce
9096	progressive frames.
9097
9098	The description of the accepted parameters follows.
9099
9100	@table @option
9101	@item thresh
9102	Set the threshold which affects the filter's tolerance when
9103	determining if a pixel line must be processed. It must be an integer
9104	in the range [0,255] and defaults to 10. A value of 0 will result in
9105	applying the process on every pixels.
9106
9107	@item map
9108	Paint pixels exceeding the threshold value to white if set to 1.
9109	Default is 0.
9110
9111	@item order
9112	Set the fields order. Swap fields if set to 1, leave fields alone if
9113	0. Default is 0.
9114
9115	@item sharp
9116	Enable additional sharpening if set to 1. Default is 0.
9117
9118	@item twoway
9119	Enable twoway sharpening if set to 1. Default is 0.
9120	@end table
9121
9122	@subsection Examples
9123
9124	@itemize
9125	@item
9126	Apply default values:
9127	@example
9128	kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
9129	@end example
9130
9131	@item
9132	Enable additional sharpening:
9133	@example
9134	kerndeint=sharp=1
9135	@end example
9136
9137	@item
9138	Paint processed pixels in white:
9139	@example
9140	kerndeint=map=1
9141	@end example
9142	@end itemize
9143
9144	@section lenscorrection
9145
9146	Correct radial lens distortion
9147
9148	This filter can be used to correct for radial distortion as can result from the use
9149	of wide angle lenses, and thereby re-rectify the image. To find the right parameters
9150	one can use tools available for example as part of opencv or simply trial-and-error.
9151	To use opencv use the calibration sample (under samples/cpp) from the opencv sources
9152	and extract the k1 and k2 coefficients from the resulting matrix.
9153
9154	Note that effectively the same filter is available in the open-source tools Krita and
9155	Digikam from the KDE project.
9156
9157	In contrast to the @ref{vignette} filter, which can also be used to compensate lens errors,
9158	this filter corrects the distortion of the image, whereas @ref{vignette} corrects the
9159	brightness distribution, so you may want to use both filters together in certain
9160	cases, though you will have to take care of ordering, i.e. whether vignetting should
9161	be applied before or after lens correction.
9162
9163	@subsection Options
9164
9165	The filter accepts the following options:
9166
9167	@table @option
9168	@item cx
9169	Relative x-coordinate of the focal point of the image, and thereby the center of the
9170	distortion. This value has a range [0,1] and is expressed as fractions of the image
9171	width.
9172	@item cy
9173	Relative y-coordinate of the focal point of the image, and thereby the center of the
9174	distortion. This value has a range [0,1] and is expressed as fractions of the image
9175	height.
9176	@item k1
9177	Coefficient of the quadratic correction term. 0.5 means no correction.
9178	@item k2
9179	Coefficient of the double quadratic correction term. 0.5 means no correction.
9180	@end table
9181
9182	The formula that generates the correction is:
9183
9184	@var{r_src} = @var{r_tgt} * (1 + @var{k1} * (@var{r_tgt} / @var{r_0})^2 + @var{k2} * (@var{r_tgt} / @var{r_0})^4)
9185
9186	where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the
9187	distances from the focal point in the source and target images, respectively.
9188
9189	@section loop
9190
9191	Loop video frames.
9192
9193	The filter accepts the following options:
9194
9195	@table @option
9196	@item loop
9197	Set the number of loops.
9198
9199	@item size
9200	Set maximal size in number of frames.
9201
9202	@item start
9203	Set first frame of loop.
9204	@end table
9205
9206	@anchor{lut3d}
9207	@section lut3d
9208
9209	Apply a 3D LUT to an input video.
9210
9211	The filter accepts the following options:
9212
9213	@table @option
9214	@item file
9215	Set the 3D LUT file name.
9216
9217	Currently supported formats:
9218	@table @samp
9219	@item 3dl
9220	AfterEffects
9221	@item cube
9222	Iridas
9223	@item dat
9224	DaVinci
9225	@item m3d
9226	Pandora
9227	@end table
9228	@item interp
9229	Select interpolation mode.
9230
9231	Available values are:
9232
9233	@table @samp
9234	@item nearest
9235	Use values from the nearest defined point.
9236	@item trilinear
9237	Interpolate values using the 8 points defining a cube.
9238	@item tetrahedral
9239	Interpolate values using a tetrahedron.
9240	@end table
9241	@end table
9242
9243	@section lut, lutrgb, lutyuv
9244
9245	Compute a look-up table for binding each pixel component input value
9246	to an output value, and apply it to the input video.
9247
9248	@var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
9249	to an RGB input video.
9250
9251	These filters accept the following parameters:
9252	@table @option
9253	@item c0
9254	set first pixel component expression
9255	@item c1
9256	set second pixel component expression
9257	@item c2
9258	set third pixel component expression
9259	@item c3
9260	set fourth pixel component expression, corresponds to the alpha component
9261
9262	@item r
9263	set red component expression
9264	@item g
9265	set green component expression
9266	@item b
9267	set blue component expression
9268	@item a
9269	alpha component expression
9270
9271	@item y
9272	set Y/luminance component expression
9273	@item u
9274	set U/Cb component expression
9275	@item v
9276	set V/Cr component expression
9277	@end table
9278
9279	Each of them specifies the expression to use for computing the lookup table for
9280	the corresponding pixel component values.
9281
9282	The exact component associated to each of the @var{c*} options depends on the
9283	format in input.
9284
9285	The @var{lut} filter requires either YUV or RGB pixel formats in input,
9286	@var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
9287
9288	The expressions can contain the following constants and functions:
9289
9290	@table @option
9291	@item w
9292	@item h
9293	The input width and height.
9294
9295	@item val
9296	The input value for the pixel component.
9297
9298	@item clipval
9299	The input value, clipped to the @var{minval}-@var{maxval} range.
9300
9301	@item maxval
9302	The maximum value for the pixel component.
9303
9304	@item minval
9305	The minimum value for the pixel component.
9306
9307	@item negval
9308	The negated value for the pixel component value, clipped to the
9309	@var{minval}-@var{maxval} range; it corresponds to the expression
9310	"maxval-clipval+minval".
9311
9312	@item clip(val)
9313	The computed value in @var{val}, clipped to the
9314	@var{minval}-@var{maxval} range.
9315
9316	@item gammaval(gamma)
9317	The computed gamma correction value of the pixel component value,
9318	clipped to the @var{minval}-@var{maxval} range. It corresponds to the
9319	expression
9320	"pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
9321
9322	@end table
9323
9324	All expressions default to "val".
9325
9326	@subsection Examples
9327
9328	@itemize
9329	@item
9330	Negate input video:
9331	@example
9332	lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
9333	lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
9334	@end example
9335
9336	The above is the same as:
9337	@example
9338	lutrgb="r=negval:g=negval:b=negval"
9339	lutyuv="y=negval:u=negval:v=negval"
9340	@end example
9341
9342	@item
9343	Negate luminance:
9344	@example
9345	lutyuv=y=negval
9346	@end example
9347
9348	@item
9349	Remove chroma components, turning the video into a graytone image:
9350	@example
9351	lutyuv="u=128:v=128"
9352	@end example
9353
9354	@item
9355	Apply a luma burning effect:
9356	@example
9357	lutyuv="y=2*val"
9358	@end example
9359
9360	@item
9361	Remove green and blue components:
9362	@example
9363	lutrgb="g=0:b=0"
9364	@end example
9365
9366	@item
9367	Set a constant alpha channel value on input:
9368	@example
9369	format=rgba,lutrgb=a="maxval-minval/2"
9370	@end example
9371
9372	@item
9373	Correct luminance gamma by a factor of 0.5:
9374	@example
9375	lutyuv=y=gammaval(0.5)
9376	@end example
9377
9378	@item
9379	Discard least significant bits of luma:
9380	@example
9381	lutyuv=y='bitand(val, 128+64+32)'
9382	@end example
9383
9384	@item
9385	Technicolor like effect:
9386	@example
9387	lutyuv=u='(val-maxval/2)*2+maxval/2':v='(val-maxval/2)*2+maxval/2'
9388	@end example
9389	@end itemize
9390
9391	@section lut2
9392
9393	Compute and apply a lookup table from two video inputs.
9394
9395	This filter accepts the following parameters:
9396	@table @option
9397	@item c0
9398	set first pixel component expression
9399	@item c1
9400	set second pixel component expression
9401	@item c2
9402	set third pixel component expression
9403	@item c3
9404	set fourth pixel component expression, corresponds to the alpha component
9405	@end table
9406
9407	Each of them specifies the expression to use for computing the lookup table for
9408	the corresponding pixel component values.
9409
9410	The exact component associated to each of the @var{c*} options depends on the
9411	format in inputs.
9412
9413	The expressions can contain the following constants:
9414
9415	@table @option
9416	@item w
9417	@item h
9418	The input width and height.
9419
9420	@item x
9421	The first input value for the pixel component.
9422
9423	@item y
9424	The second input value for the pixel component.
9425
9426	@item bdx
9427	The first input video bit depth.
9428
9429	@item bdy
9430	The second input video bit depth.
9431	@end table
9432
9433	All expressions default to "x".
9434
9435	@subsection Examples
9436
9437	@itemize
9438	@item
9439	Highlight differences between two RGB video streams:
9440	@example
9441	lut2='ifnot(x-y,0,pow(2,bdx)-1):ifnot(x-y,0,pow(2,bdx)-1):ifnot(x-y,0,pow(2,bdx)-1)'
9442	@end example
9443
9444	@item
9445	Highlight differences between two YUV video streams:
9446	@example
9447	lut2='ifnot(x-y,0,pow(2,bdx)-1):ifnot(x-y,pow(2,bdx-1),pow(2,bdx)-1):ifnot(x-y,pow(2,bdx-1),pow(2,bdx)-1)'
9448	@end example
9449	@end itemize
9450
9451	@section maskedclamp
9452
9453	Clamp the first input stream with the second input and third input stream.
9454
9455	Returns the value of first stream to be between second input
9456	stream - @code{undershoot} and third input stream + @code{overshoot}.
9457
9458	This filter accepts the following options:
9459	@table @option
9460	@item undershoot
9461	Default value is @code{0}.
9462
9463	@item overshoot
9464	Default value is @code{0}.
9465
9466	@item planes
9467	Set which planes will be processed as bitmap, unprocessed planes will be
9468	copied from first stream.
9469	By default value 0xf, all planes will be processed.
9470	@end table
9471
9472	@section maskedmerge
9473
9474	Merge the first input stream with the second input stream using per pixel
9475	weights in the third input stream.
9476
9477	A value of 0 in the third stream pixel component means that pixel component
9478	from first stream is returned unchanged, while maximum value (eg. 255 for
9479	8-bit videos) means that pixel component from second stream is returned
9480	unchanged. Intermediate values define the amount of merging between both
9481	input stream's pixel components.
9482
9483	This filter accepts the following options:
9484	@table @option
9485	@item planes
9486	Set which planes will be processed as bitmap, unprocessed planes will be
9487	copied from first stream.
9488	By default value 0xf, all planes will be processed.
9489	@end table
9490
9491	@section mcdeint
9492
9493	Apply motion-compensation deinterlacing.
9494
9495	It needs one field per frame as input and must thus be used together
9496	with yadif=1/3 or equivalent.
9497
9498	This filter accepts the following options:
9499	@table @option
9500	@item mode
9501	Set the deinterlacing mode.
9502
9503	It accepts one of the following values:
9504	@table @samp
9505	@item fast
9506	@item medium
9507	@item slow
9508	use iterative motion estimation
9509	@item extra_slow
9510	like @samp{slow}, but use multiple reference frames.
9511	@end table
9512	Default value is @samp{fast}.
9513
9514	@item parity
9515	Set the picture field parity assumed for the input video. It must be
9516	one of the following values:
9517
9518	@table @samp
9519	@item 0, tff
9520	assume top field first
9521	@item 1, bff
9522	assume bottom field first
9523	@end table
9524
9525	Default value is @samp{bff}.
9526
9527	@item qp
9528	Set per-block quantization parameter (QP) used by the internal
9529	encoder.
9530
9531	Higher values should result in a smoother motion vector field but less
9532	optimal individual vectors. Default value is 1.
9533	@end table
9534
9535	@section mergeplanes
9536
9537	Merge color channel components from several video streams.
9538
9539	The filter accepts up to 4 input streams, and merge selected input
9540	planes to the output video.
9541
9542	This filter accepts the following options:
9543	@table @option
9544	@item mapping
9545	Set input to output plane mapping. Default is @code{0}.
9546
9547	The mappings is specified as a bitmap. It should be specified as a
9548	hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
9549	mapping for the first plane of the output stream. 'A' sets the number of
9550	the input stream to use (from 0 to 3), and 'a' the plane number of the
9551	corresponding input to use (from 0 to 3). The rest of the mappings is
9552	similar, 'Bb' describes the mapping for the output stream second
9553	plane, 'Cc' describes the mapping for the output stream third plane and
9554	'Dd' describes the mapping for the output stream fourth plane.
9555
9556	@item format
9557	Set output pixel format. Default is @code{yuva444p}.
9558	@end table
9559
9560	@subsection Examples
9561
9562	@itemize
9563	@item
9564	Merge three gray video streams of same width and height into single video stream:
9565	@example
9566	[a0][a1][a2]mergeplanes=0x001020:yuv444p
9567	@end example
9568
9569	@item
9570	Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
9571	@example
9572	[a0][a1]mergeplanes=0x00010210:yuva444p
9573	@end example
9574
9575	@item
9576	Swap Y and A plane in yuva444p stream:
9577	@example
9578	format=yuva444p,mergeplanes=0x03010200:yuva444p
9579	@end example
9580
9581	@item
9582	Swap U and V plane in yuv420p stream:
9583	@example
9584	format=yuv420p,mergeplanes=0x000201:yuv420p
9585	@end example
9586
9587	@item
9588	Cast a rgb24 clip to yuv444p:
9589	@example
9590	format=rgb24,mergeplanes=0x000102:yuv444p
9591	@end example
9592	@end itemize
9593
9594	@section mestimate
9595
9596	Estimate and export motion vectors using block matching algorithms.
9597	Motion vectors are stored in frame side data to be used by other filters.
9598
9599	This filter accepts the following options:
9600	@table @option
9601	@item method
9602	Specify the motion estimation method. Accepts one of the following values:
9603
9604	@table @samp
9605	@item esa
9606	Exhaustive search algorithm.
9607	@item tss
9608	Three step search algorithm.
9609	@item tdls
9610	Two dimensional logarithmic search algorithm.
9611	@item ntss
9612	New three step search algorithm.
9613	@item fss
9614	Four step search algorithm.
9615	@item ds
9616	Diamond search algorithm.
9617	@item hexbs
9618	Hexagon-based search algorithm.
9619	@item epzs
9620	Enhanced predictive zonal search algorithm.
9621	@item umh
9622	Uneven multi-hexagon search algorithm.
9623	@end table
9624	Default value is @samp{esa}.
9625
9626	@item mb_size
9627	Macroblock size. Default @code{16}.
9628
9629	@item search_param
9630	Search parameter. Default @code{7}.
9631	@end table
9632
9633	@section midequalizer
9634
9635	Apply Midway Image Equalization effect using two video streams.
9636
9637	Midway Image Equalization adjusts a pair of images to have the same
9638	histogram, while maintaining their dynamics as much as possible. It's
9639	useful for e.g. matching exposures from a pair of stereo cameras.
9640
9641	This filter has two inputs and one output, which must be of same pixel format, but
9642	may be of different sizes. The output of filter is first input adjusted with
9643	midway histogram of both inputs.
9644
9645	This filter accepts the following option:
9646
9647	@table @option
9648	@item planes
9649	Set which planes to process. Default is @code{15}, which is all available planes.
9650	@end table
9651
9652	@section minterpolate
9653
9654	Convert the video to specified frame rate using motion interpolation.
9655
9656	This filter accepts the following options:
9657	@table @option
9658	@item fps
9659	Specify the output frame rate. This can be rational e.g. @code{60000/1001}. Frames are dropped if @var{fps} is lower than source fps. Default @code{60}.
9660
9661	@item mi_mode
9662	Motion interpolation mode. Following values are accepted:
9663	@table @samp
9664	@item dup
9665	Duplicate previous or next frame for interpolating new ones.
9666	@item blend
9667	Blend source frames. Interpolated frame is mean of previous and next frames.
9668	@item mci
9669	Motion compensated interpolation. Following options are effective when this mode is selected:
9670
9671	@table @samp
9672	@item mc_mode
9673	Motion compensation mode. Following values are accepted:
9674	@table @samp
9675	@item obmc
9676	Overlapped block motion compensation.
9677	@item aobmc
9678	Adaptive overlapped block motion compensation. Window weighting coefficients are controlled adaptively according to the reliabilities of the neighboring motion vectors to reduce oversmoothing.
9679	@end table
9680	Default mode is @samp{obmc}.
9681
9682	@item me_mode
9683	Motion estimation mode. Following values are accepted:
9684	@table @samp
9685	@item bidir
9686	Bidirectional motion estimation. Motion vectors are estimated for each source frame in both forward and backward directions.
9687	@item bilat
9688	Bilateral motion estimation. Motion vectors are estimated directly for interpolated frame.
9689	@end table
9690	Default mode is @samp{bilat}.
9691
9692	@item me
9693	The algorithm to be used for motion estimation. Following values are accepted:
9694	@table @samp
9695	@item esa
9696	Exhaustive search algorithm.
9697	@item tss
9698	Three step search algorithm.
9699	@item tdls
9700	Two dimensional logarithmic search algorithm.
9701	@item ntss
9702	New three step search algorithm.
9703	@item fss
9704	Four step search algorithm.
9705	@item ds
9706	Diamond search algorithm.
9707	@item hexbs
9708	Hexagon-based search algorithm.
9709	@item epzs
9710	Enhanced predictive zonal search algorithm.
9711	@item umh
9712	Uneven multi-hexagon search algorithm.
9713	@end table
9714	Default algorithm is @samp{epzs}.
9715
9716	@item mb_size
9717	Macroblock size. Default @code{16}.
9718
9719	@item search_param
9720	Motion estimation search parameter. Default @code{32}.
9721
9722	@item vsbmc
9723	Enable variable-size block motion compensation. Motion estimation is applied with smaller block sizes at object boundaries in order to make the them less blur. Default is @code{0} (disabled).
9724	@end table
9725	@end table
9726
9727	@item scd
9728	Scene change detection method. Scene change leads motion vectors to be in random direction. Scene change detection replace interpolated frames by duplicate ones. May not be needed for other modes. Following values are accepted:
9729	@table @samp
9730	@item none
9731	Disable scene change detection.
9732	@item fdiff
9733	Frame difference. Corresponding pixel values are compared and if it satisfies @var{scd_threshold} scene change is detected.
9734	@end table
9735	Default method is @samp{fdiff}.
9736
9737	@item scd_threshold
9738	Scene change detection threshold. Default is @code{5.0}.
9739	@end table
9740
9741	@section mpdecimate
9742
9743	Drop frames that do not differ greatly from the previous frame in
9744	order to reduce frame rate.
9745
9746	The main use of this filter is for very-low-bitrate encoding
9747	(e.g. streaming over dialup modem), but it could in theory be used for
9748	fixing movies that were inverse-telecined incorrectly.
9749
9750	A description of the accepted options follows.
9751
9752	@table @option
9753	@item max
9754	Set the maximum number of consecutive frames which can be dropped (if
9755	positive), or the minimum interval between dropped frames (if
9756	negative). If the value is 0, the frame is dropped unregarding the
9757	number of previous sequentially dropped frames.
9758
9759	Default value is 0.
9760
9761	@item hi
9762	@item lo
9763	@item frac
9764	Set the dropping threshold values.
9765
9766	Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
9767	represent actual pixel value differences, so a threshold of 64
9768	corresponds to 1 unit of difference for each pixel, or the same spread
9769	out differently over the block.
9770
9771	A frame is a candidate for dropping if no 8x8 blocks differ by more
9772	than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
9773	meaning the whole image) differ by more than a threshold of @option{lo}.
9774
9775	Default value for @option{hi} is 64*12, default value for @option{lo} is
9776	64*5, and default value for @option{frac} is 0.33.
9777	@end table
9778
9779
9780	@section negate
9781
9782	Negate input video.
9783
9784	It accepts an integer in input; if non-zero it negates the
9785	alpha component (if available). The default value in input is 0.
9786
9787	@section nlmeans
9788
9789	Denoise frames using Non-Local Means algorithm.
9790
9791	Each pixel is adjusted by looking for other pixels with similar contexts. This
9792	context similarity is defined by comparing their surrounding patches of size
9793	@option{p}x@option{p}. Patches are searched in an area of @option{r}x@option{r}
9794	around the pixel.
9795
9796	Note that the research area defines centers for patches, which means some
9797	patches will be made of pixels outside that research area.
9798
9799	The filter accepts the following options.
9800
9801	@table @option
9802	@item s
9803	Set denoising strength.
9804
9805	@item p
9806	Set patch size.
9807
9808	@item pc
9809	Same as @option{p} but for chroma planes.
9810
9811	The default value is @var{0} and means automatic.
9812
9813	@item r
9814	Set research size.
9815
9816	@item rc
9817	Same as @option{r} but for chroma planes.
9818
9819	The default value is @var{0} and means automatic.
9820	@end table
9821
9822	@section nnedi
9823
9824	Deinterlace video using neural network edge directed interpolation.
9825
9826	This filter accepts the following options:
9827
9828	@table @option
9829	@item weights
9830	Mandatory option, without binary file filter can not work.
9831	Currently file can be found here:
9832	https://github.com/dubhater/vapoursynth-nnedi3/blob/master/src/nnedi3_weights.bin
9833
9834	@item deint
9835	Set which frames to deinterlace, by default it is @code{all}.
9836	Can be @code{all} or @code{interlaced}.
9837
9838	@item field
9839	Set mode of operation.
9840
9841	Can be one of the following:
9842
9843	@table @samp
9844	@item af
9845	Use frame flags, both fields.
9846	@item a
9847	Use frame flags, single field.
9848	@item t
9849	Use top field only.
9850	@item b
9851	Use bottom field only.
9852	@item tf
9853	Use both fields, top first.
9854	@item bf
9855	Use both fields, bottom first.
9856	@end table
9857
9858	@item planes
9859	Set which planes to process, by default filter process all frames.
9860
9861	@item nsize
9862	Set size of local neighborhood around each pixel, used by the predictor neural
9863	network.
9864
9865	Can be one of the following:
9866
9867	@table @samp
9868	@item s8x6
9869	@item s16x6
9870	@item s32x6
9871	@item s48x6
9872	@item s8x4
9873	@item s16x4
9874	@item s32x4
9875	@end table
9876
9877	@item nns
9878	Set the number of neurons in predicctor neural network.
9879	Can be one of the following:
9880
9881	@table @samp
9882	@item n16
9883	@item n32
9884	@item n64
9885	@item n128
9886	@item n256
9887	@end table
9888
9889	@item qual
9890	Controls the number of different neural network predictions that are blended
9891	together to compute the final output value. Can be @code{fast}, default or
9892	@code{slow}.
9893
9894	@item etype
9895	Set which set of weights to use in the predictor.
9896	Can be one of the following:
9897
9898	@table @samp
9899	@item a
9900	weights trained to minimize absolute error
9901	@item s
9902	weights trained to minimize squared error
9903	@end table
9904
9905	@item pscrn
9906	Controls whether or not the prescreener neural network is used to decide
9907	which pixels should be processed by the predictor neural network and which
9908	can be handled by simple cubic interpolation.
9909	The prescreener is trained to know whether cubic interpolation will be
9910	sufficient for a pixel or whether it should be predicted by the predictor nn.
9911	The computational complexity of the prescreener nn is much less than that of
9912	the predictor nn. Since most pixels can be handled by cubic interpolation,
9913	using the prescreener generally results in much faster processing.
9914	The prescreener is pretty accurate, so the difference between using it and not
9915	using it is almost always unnoticeable.
9916
9917	Can be one of the following:
9918
9919	@table @samp
9920	@item none
9921	@item original
9922	@item new
9923	@end table
9924
9925	Default is @code{new}.
9926
9927	@item fapprox
9928	Set various debugging flags.
9929	@end table
9930
9931	@section noformat
9932
9933	Force libavfilter not to use any of the specified pixel formats for the
9934	input to the next filter.
9935
9936	It accepts the following parameters:
9937	@table @option
9938
9939	@item pix_fmts
9940	A '|'-separated list of pixel format names, such as
9941	apix_fmts=yuv420p|monow|rgb24".
9942
9943	@end table
9944
9945	@subsection Examples
9946
9947	@itemize
9948	@item
9949	Force libavfilter to use a format different from @var{yuv420p} for the
9950	input to the vflip filter:
9951	@example
9952	noformat=pix_fmts=yuv420p,vflip
9953	@end example
9954
9955	@item
9956	Convert the input video to any of the formats not contained in the list:
9957	@example
9958	noformat=yuv420p|yuv444p|yuv410p
9959	@end example
9960	@end itemize
9961
9962	@section noise
9963
9964	Add noise on video input frame.
9965
9966	The filter accepts the following options:
9967
9968	@table @option
9969	@item all_seed
9970	@item c0_seed
9971	@item c1_seed
9972	@item c2_seed
9973	@item c3_seed
9974	Set noise seed for specific pixel component or all pixel components in case
9975	of @var{all_seed}. Default value is @code{123457}.
9976
9977	@item all_strength, alls
9978	@item c0_strength, c0s
9979	@item c1_strength, c1s
9980	@item c2_strength, c2s
9981	@item c3_strength, c3s
9982	Set noise strength for specific pixel component or all pixel components in case
9983	@var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
9984
9985	@item all_flags, allf
9986	@item c0_flags, c0f
9987	@item c1_flags, c1f
9988	@item c2_flags, c2f
9989	@item c3_flags, c3f
9990	Set pixel component flags or set flags for all components if @var{all_flags}.
9991	Available values for component flags are:
9992	@table @samp
9993	@item a
9994	averaged temporal noise (smoother)
9995	@item p
9996	mix random noise with a (semi)regular pattern
9997	@item t
9998	temporal noise (noise pattern changes between frames)
9999	@item u
10000	uniform noise (gaussian otherwise)
10001	@end table
10002	@end table
10003
10004	@subsection Examples
10005
10006	Add temporal and uniform noise to input video:
10007	@example
10008	noise=alls=20:allf=t+u
10009	@end example
10010
10011	@section null
10012
10013	Pass the video source unchanged to the output.
10014
10015	@section ocr
10016	Optical Character Recognition
10017
10018	This filter uses Tesseract for optical character recognition.
10019
10020	It accepts the following options:
10021
10022	@table @option
10023	@item datapath
10024	Set datapath to tesseract data. Default is to use whatever was
10025	set at installation.
10026
10027	@item language
10028	Set language, default is "eng".
10029
10030	@item whitelist
10031	Set character whitelist.
10032
10033	@item blacklist
10034	Set character blacklist.
10035	@end table
10036
10037	The filter exports recognized text as the frame metadata @code{lavfi.ocr.text}.
10038
10039	@section ocv
10040
10041	Apply a video transform using libopencv.
10042
10043	To enable this filter, install the libopencv library and headers and
10044	configure FFmpeg with @code{--enable-libopencv}.
10045
10046	It accepts the following parameters:
10047
10048	@table @option
10049
10050	@item filter_name
10051	The name of the libopencv filter to apply.
10052
10053	@item filter_params
10054	The parameters to pass to the libopencv filter. If not specified, the default
10055	values are assumed.
10056
10057	@end table
10058
10059	Refer to the official libopencv documentation for more precise
10060	information:
10061	@url{http://docs.opencv.org/master/modules/imgproc/doc/filtering.html}
10062
10063	Several libopencv filters are supported; see the following subsections.
10064
10065	@anchor{dilate}
10066	@subsection dilate
10067
10068	Dilate an image by using a specific structuring element.
10069	It corresponds to the libopencv function @code{cvDilate}.
10070
10071	It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
10072
10073	@var{struct_el} represents a structuring element, and has the syntax:
10074	@var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
10075
10076	@var{cols} and @var{rows} represent the number of columns and rows of
10077	the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
10078	point, and @var{shape} the shape for the structuring element. @var{shape}
10079	must be "rect", "cross", "ellipse", or "custom".
10080
10081	If the value for @var{shape} is "custom", it must be followed by a
10082	string of the form "=@var{filename}". The file with name
10083	@var{filename} is assumed to represent a binary image, with each
10084	printable character corresponding to a bright pixel. When a custom
10085	@var{shape} is used, @var{cols} and @var{rows} are ignored, the number
10086	or columns and rows of the read file are assumed instead.
10087
10088	The default value for @var{struct_el} is "3x3+0x0/rect".
10089
10090	@var{nb_iterations} specifies the number of times the transform is
10091	applied to the image, and defaults to 1.
10092
10093	Some examples:
10094	@example
10095	# Use the default values
10096	ocv=dilate
10097
10098	# Dilate using a structuring element with a 5x5 cross, iterating two times
10099	ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
10100
10101	# Read the shape from the file diamond.shape, iterating two times.
10102	# The file diamond.shape may contain a pattern of characters like this
10103	# *
10104	# ***
10105	# *****
10106	# ***
10107	# *
10108	# The specified columns and rows are ignored
10109	# but the anchor point coordinates are not
10110	ocv=dilate:0x0+2x2/custom=diamond.shape|2
10111	@end example
10112
10113	@subsection erode
10114
10115	Erode an image by using a specific structuring element.
10116	It corresponds to the libopencv function @code{cvErode}.
10117
10118	It accepts the parameters: @var{struct_el}:@var{nb_iterations},
10119	with the same syntax and semantics as the @ref{dilate} filter.
10120
10121	@subsection smooth
10122
10123	Smooth the input video.
10124
10125	The filter takes the following parameters:
10126	@var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
10127
10128	@var{type} is the type of smooth filter to apply, and must be one of
10129	the following values: "blur", "blur_no_scale", "median", "gaussian",
10130	or "bilateral". The default value is "gaussian".
10131
10132	The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
10133	depend on the smooth type. @var{param1} and
10134	@var{param2} accept integer positive values or 0. @var{param3} and
10135	@var{param4} accept floating point values.
10136
10137	The default value for @var{param1} is 3. The default value for the
10138	other parameters is 0.
10139
10140	These parameters correspond to the parameters assigned to the
10141	libopencv function @code{cvSmooth}.
10142
10143	@anchor{overlay}
10144	@section overlay
10145
10146	Overlay one video on top of another.
10147
10148	It takes two inputs and has one output. The first input is the "main"
10149	video on which the second input is overlaid.
10150
10151	It accepts the following parameters:
10152
10153	A description of the accepted options follows.
10154
10155	@table @option
10156	@item x
10157	@item y
10158	Set the expression for the x and y coordinates of the overlaid video
10159	on the main video. Default value is "0" for both expressions. In case
10160	the expression is invalid, it is set to a huge value (meaning that the
10161	overlay will not be displayed within the output visible area).
10162
10163	@item eof_action
10164	The action to take when EOF is encountered on the secondary input; it accepts
10165	one of the following values:
10166
10167	@table @option
10168	@item repeat
10169	Repeat the last frame (the default).
10170	@item endall
10171	End both streams.
10172	@item pass
10173	Pass the main input through.
10174	@end table
10175
10176	@item eval
10177	Set when the expressions for @option{x}, and @option{y} are evaluated.
10178
10179	It accepts the following values:
10180	@table @samp
10181	@item init
10182	only evaluate expressions once during the filter initialization or
10183	when a command is processed
10184
10185	@item frame
10186	evaluate expressions for each incoming frame
10187	@end table
10188
10189	Default value is @samp{frame}.
10190
10191	@item shortest
10192	If set to 1, force the output to terminate when the shortest input
10193	terminates. Default value is 0.
10194
10195	@item format
10196	Set the format for the output video.
10197
10198	It accepts the following values:
10199	@table @samp
10200	@item yuv420
10201	force YUV420 output
10202
10203	@item yuv422
10204	force YUV422 output
10205
10206	@item yuv444
10207	force YUV444 output
10208
10209	@item rgb
10210	force packed RGB output
10211
10212	@item gbrp
10213	force planar RGB output
10214	@end table
10215
10216	Default value is @samp{yuv420}.
10217
10218	@item rgb @emph{(deprecated)}
10219	If set to 1, force the filter to accept inputs in the RGB
10220	color space. Default value is 0. This option is deprecated, use
10221	@option{format} instead.
10222
10223	@item repeatlast
10224	If set to 1, force the filter to draw the last overlay frame over the
10225	main input until the end of the stream. A value of 0 disables this
10226	behavior. Default value is 1.
10227	@end table
10228
10229	The @option{x}, and @option{y} expressions can contain the following
10230	parameters.
10231
10232	@table @option
10233	@item main_w, W
10234	@item main_h, H
10235	The main input width and height.
10236
10237	@item overlay_w, w
10238	@item overlay_h, h
10239	The overlay input width and height.
10240
10241	@item x
10242	@item y
10243	The computed values for @var{x} and @var{y}. They are evaluated for
10244	each new frame.
10245
10246	@item hsub
10247	@item vsub
10248	horizontal and vertical chroma subsample values of the output
10249	format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
10250	@var{vsub} is 1.
10251
10252	@item n
10253	the number of input frame, starting from 0
10254
10255	@item pos
10256	the position in the file of the input frame, NAN if unknown
10257
10258	@item t
10259	The timestamp, expressed in seconds. It's NAN if the input timestamp is unknown.
10260
10261	@end table
10262
10263	Note that the @var{n}, @var{pos}, @var{t} variables are available only
10264	when evaluation is done @emph{per frame}, and will evaluate to NAN
10265	when @option{eval} is set to @samp{init}.
10266
10267	Be aware that frames are taken from each input video in timestamp
10268	order, hence, if their initial timestamps differ, it is a good idea
10269	to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
10270	have them begin in the same zero timestamp, as the example for
10271	the @var{movie} filter does.
10272
10273	You can chain together more overlays but you should test the
10274	efficiency of such approach.
10275
10276	@subsection Commands
10277
10278	This filter supports the following commands:
10279	@table @option
10280	@item x
10281	@item y
10282	Modify the x and y of the overlay input.
10283	The command accepts the same syntax of the corresponding option.
10284
10285	If the specified expression is not valid, it is kept at its current
10286	value.
10287	@end table
10288
10289	@subsection Examples
10290
10291	@itemize
10292	@item
10293	Draw the overlay at 10 pixels from the bottom right corner of the main
10294	video:
10295	@example
10296	overlay=main_w-overlay_w-10:main_h-overlay_h-10
10297	@end example
10298
10299	Using named options the example above becomes:
10300	@example
10301	overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
10302	@end example
10303
10304	@item
10305	Insert a transparent PNG logo in the bottom left corner of the input,
10306	using the @command{ffmpeg} tool with the @code{-filter_complex} option:
10307	@example
10308	ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
10309	@end example
10310
10311	@item
10312	Insert 2 different transparent PNG logos (second logo on bottom
10313	right corner) using the @command{ffmpeg} tool:
10314	@example
10315	ffmpeg -i input -i logo1 -i logo2 -filter_complex 'overlay=x=10:y=H-h-10,overlay=x=W-w-10:y=H-h-10' output
10316	@end example
10317
10318	@item
10319	Add a transparent color layer on top of the main video; @code{WxH}
10320	must specify the size of the main input to the overlay filter:
10321	@example
10322	color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
10323	@end example
10324
10325	@item
10326	Play an original video and a filtered version (here with the deshake
10327	filter) side by side using the @command{ffplay} tool:
10328	@example
10329	ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
10330	@end example
10331
10332	The above command is the same as:
10333	@example
10334	ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
10335	@end example
10336
10337	@item
10338	Make a sliding overlay appearing from the left to the right top part of the
10339	screen starting since time 2:
10340	@example
10341	overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
10342	@end example
10343
10344	@item
10345	Compose output by putting two input videos side to side:
10346	@example
10347	ffmpeg -i left.avi -i right.avi -filter_complex "
10348	nullsrc=size=200x100 [background];
10349	[0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
10350	[1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
10351	[background][left] overlay=shortest=1 [background+left];
10352	[background+left][right] overlay=shortest=1:x=100 [left+right]
10353	"
10354	@end example
10355
10356	@item
10357	Mask 10-20 seconds of a video by applying the delogo filter to a section
10358	@example
10359	ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
10360	-vf '[in]split[split_main][split_delogo];[split_delogo]trim=start=360:end=371,delogo=0:0:640:480[delogoed];[split_main][delogoed]overlay=eof_action=pass[out]'
10361	masked.avi
10362	@end example
10363
10364	@item
10365	Chain several overlays in cascade:
10366	@example
10367	nullsrc=s=200x200 [bg];
10368	testsrc=s=100x100, split=4 [in0][in1][in2][in3];
10369	[in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
10370	[in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
10371	[in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
10372	[in3] null, [mid2] overlay=100:100 [out0]
10373	@end example
10374
10375	@end itemize
10376
10377	@section owdenoise
10378
10379	Apply Overcomplete Wavelet denoiser.
10380
10381	The filter accepts the following options:
10382
10383	@table @option
10384	@item depth
10385	Set depth.
10386
10387	Larger depth values will denoise lower frequency components more, but
10388	slow down filtering.
10389
10390	Must be an int in the range 8-16, default is @code{8}.
10391
10392	@item luma_strength, ls
10393	Set luma strength.
10394
10395	Must be a double value in the range 0-1000, default is @code{1.0}.
10396
10397	@item chroma_strength, cs
10398	Set chroma strength.
10399
10400	Must be a double value in the range 0-1000, default is @code{1.0}.
10401	@end table
10402
10403	@anchor{pad}
10404	@section pad
10405
10406	Add paddings to the input image, and place the original input at the
10407	provided @var{x}, @var{y} coordinates.
10408
10409	It accepts the following parameters:
10410
10411	@table @option
10412	@item width, w
10413	@item height, h
10414	Specify an expression for the size of the output image with the
10415	paddings added. If the value for @var{width} or @var{height} is 0, the
10416	corresponding input size is used for the output.
10417
10418	The @var{width} expression can reference the value set by the
10419	@var{height} expression, and vice versa.
10420
10421	The default value of @var{width} and @var{height} is 0.
10422
10423	@item x
10424	@item y
10425	Specify the offsets to place the input image at within the padded area,
10426	with respect to the top/left border of the output image.
10427
10428	The @var{x} expression can reference the value set by the @var{y}
10429	expression, and vice versa.
10430
10431	The default value of @var{x} and @var{y} is 0.
10432
10433	@item color
10434	Specify the color of the padded area. For the syntax of this option,
10435	check the "Color" section in the ffmpeg-utils manual.
10436
10437	The default value of @var{color} is "black".
10438
10439	@item eval
10440	Specify when to evaluate @var{width}, @var{height}, @var{x} and @var{y} expression.
10441
10442	It accepts the following values:
10443
10444	@table @samp
10445	@item init
10446	Only evaluate expressions once during the filter initialization or when
10447	a command is processed.
10448
10449	@item frame
10450	Evaluate expressions for each incoming frame.
10451
10452	@end table
10453
10454	Default value is @samp{init}.
10455
10456	@end table
10457
10458	The value for the @var{width}, @var{height}, @var{x}, and @var{y}
10459	options are expressions containing the following constants:
10460
10461	@table @option
10462	@item in_w
10463	@item in_h
10464	The input video width and height.
10465
10466	@item iw
10467	@item ih
10468	These are the same as @var{in_w} and @var{in_h}.
10469
10470	@item out_w
10471	@item out_h
10472	The output width and height (the size of the padded area), as
10473	specified by the @var{width} and @var{height} expressions.
10474
10475	@item ow
10476	@item oh
10477	These are the same as @var{out_w} and @var{out_h}.
10478
10479	@item x
10480	@item y
10481	The x and y offsets as specified by the @var{x} and @var{y}
10482	expressions, or NAN if not yet specified.
10483
10484	@item a
10485	same as @var{iw} / @var{ih}
10486
10487	@item sar
10488	input sample aspect ratio
10489
10490	@item dar
10491	input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
10492
10493	@item hsub
10494	@item vsub
10495	The horizontal and vertical chroma subsample values. For example for the
10496	pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
10497	@end table
10498
10499	@subsection Examples
10500
10501	@itemize
10502	@item
10503	Add paddings with the color "violet" to the input video. The output video
10504	size is 640x480, and the top-left corner of the input video is placed at
10505	column 0, row 40
10506	@example
10507	pad=640:480:0:40:violet
10508	@end example
10509
10510	The example above is equivalent to the following command:
10511	@example
10512	pad=width=640:height=480:x=0:y=40:color=violet
10513	@end example
10514
10515	@item
10516	Pad the input to get an output with dimensions increased by 3/2,
10517	and put the input video at the center of the padded area:
10518	@example
10519	pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
10520	@end example
10521
10522	@item
10523	Pad the input to get a squared output with size equal to the maximum
10524	value between the input width and height, and put the input video at
10525	the center of the padded area:
10526	@example
10527	pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
10528	@end example
10529
10530	@item
10531	Pad the input to get a final w/h ratio of 16:9:
10532	@example
10533	pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
10534	@end example
10535
10536	@item
10537	In case of anamorphic video, in order to set the output display aspect
10538	correctly, it is necessary to use @var{sar} in the expression,
10539	according to the relation:
10540	@example
10541	(ih * X / ih) * sar = output_dar
10542	X = output_dar / sar
10543	@end example
10544
10545	Thus the previous example needs to be modified to:
10546	@example
10547	pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
10548	@end example
10549
10550	@item
10551	Double the output size and put the input video in the bottom-right
10552	corner of the output padded area:
10553	@example
10554	pad="2*iw:2*ih:ow-iw:oh-ih"
10555	@end example
10556	@end itemize
10557
10558	@anchor{palettegen}
10559	@section palettegen
10560
10561	Generate one palette for a whole video stream.
10562
10563	It accepts the following options:
10564
10565	@table @option
10566	@item max_colors
10567	Set the maximum number of colors to quantize in the palette.
10568	Note: the palette will still contain 256 colors; the unused palette entries
10569	will be black.
10570
10571	@item reserve_transparent
10572	Create a palette of 255 colors maximum and reserve the last one for
10573	transparency. Reserving the transparency color is useful for GIF optimization.
10574	If not set, the maximum of colors in the palette will be 256. You probably want
10575	to disable this option for a standalone image.
10576	Set by default.
10577
10578	@item stats_mode
10579	Set statistics mode.
10580
10581	It accepts the following values:
10582	@table @samp
10583	@item full
10584	Compute full frame histograms.
10585	@item diff
10586	Compute histograms only for the part that differs from previous frame. This
10587	might be relevant to give more importance to the moving part of your input if
10588	the background is static.
10589	@item single
10590	Compute new histogram for each frame.
10591	@end table
10592
10593	Default value is @var{full}.
10594	@end table
10595
10596	The filter also exports the frame metadata @code{lavfi.color_quant_ratio}
10597	(@code{nb_color_in / nb_color_out}) which you can use to evaluate the degree of
10598	color quantization of the palette. This information is also visible at
10599	@var{info} logging level.
10600
10601	@subsection Examples
10602
10603	@itemize
10604	@item
10605	Generate a representative palette of a given video using @command{ffmpeg}:
10606	@example
10607	ffmpeg -i input.mkv -vf palettegen palette.png
10608	@end example
10609	@end itemize
10610
10611	@section paletteuse
10612
10613	Use a palette to downsample an input video stream.
10614
10615	The filter takes two inputs: one video stream and a palette. The palette must
10616	be a 256 pixels image.
10617
10618	It accepts the following options:
10619
10620	@table @option
10621	@item dither
10622	Select dithering mode. Available algorithms are:
10623	@table @samp
10624	@item bayer
10625	Ordered 8x8 bayer dithering (deterministic)
10626	@item heckbert
10627	Dithering as defined by Paul Heckbert in 1982 (simple error diffusion).
10628	Note: this dithering is sometimes considered "wrong" and is included as a
10629	reference.
10630	@item floyd_steinberg
10631	Floyd and Steingberg dithering (error diffusion)
10632	@item sierra2
10633	Frankie Sierra dithering v2 (error diffusion)
10634	@item sierra2_4a
10635	Frankie Sierra dithering v2 "Lite" (error diffusion)
10636	@end table
10637
10638	Default is @var{sierra2_4a}.
10639
10640	@item bayer_scale
10641	When @var{bayer} dithering is selected, this option defines the scale of the
10642	pattern (how much the crosshatch pattern is visible). A low value means more
10643	visible pattern for less banding, and higher value means less visible pattern
10644	at the cost of more banding.
10645
10646	The option must be an integer value in the range [0,5]. Default is @var{2}.
10647
10648	@item diff_mode
10649	If set, define the zone to process
10650
10651	@table @samp
10652	@item rectangle
10653	Only the changing rectangle will be reprocessed. This is similar to GIF
10654	cropping/offsetting compression mechanism. This option can be useful for speed
10655	if only a part of the image is changing, and has use cases such as limiting the
10656	scope of the error diffusal @option{dither} to the rectangle that bounds the
10657	moving scene (it leads to more deterministic output if the scene doesn't change
10658	much, and as a result less moving noise and better GIF compression).
10659	@end table
10660
10661	Default is @var{none}.
10662
10663	@item new
10664	Take new palette for each output frame.
10665	@end table
10666
10667	@subsection Examples
10668
10669	@itemize
10670	@item
10671	Use a palette (generated for example with @ref{palettegen}) to encode a GIF
10672	using @command{ffmpeg}:
10673	@example
10674	ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
10675	@end example
10676	@end itemize
10677
10678	@section perspective
10679
10680	Correct perspective of video not recorded perpendicular to the screen.
10681
10682	A description of the accepted parameters follows.
10683
10684	@table @option
10685	@item x0
10686	@item y0
10687	@item x1
10688	@item y1
10689	@item x2
10690	@item y2
10691	@item x3
10692	@item y3
10693	Set coordinates expression for top left, top right, bottom left and bottom right corners.
10694	Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
10695	If the @code{sense} option is set to @code{source}, then the specified points will be sent
10696	to the corners of the destination. If the @code{sense} option is set to @code{destination},
10697	then the corners of the source will be sent to the specified coordinates.
10698
10699	The expressions can use the following variables:
10700
10701	@table @option
10702	@item W
10703	@item H
10704	the width and height of video frame.
10705	@item in
10706	Input frame count.
10707	@item on
10708	Output frame count.
10709	@end table
10710
10711	@item interpolation
10712	Set interpolation for perspective correction.
10713
10714	It accepts the following values:
10715	@table @samp
10716	@item linear
10717	@item cubic
10718	@end table
10719
10720	Default value is @samp{linear}.
10721
10722	@item sense
10723	Set interpretation of coordinate options.
10724
10725	It accepts the following values:
10726	@table @samp
10727	@item 0, source
10728
10729	Send point in the source specified by the given coordinates to
10730	the corners of the destination.
10731
10732	@item 1, destination
10733
10734	Send the corners of the source to the point in the destination specified
10735	by the given coordinates.
10736
10737	Default value is @samp{source}.
10738	@end table
10739
10740	@item eval
10741	Set when the expressions for coordinates @option{x0,y0,...x3,y3} are evaluated.
10742
10743	It accepts the following values:
10744	@table @samp
10745	@item init
10746	only evaluate expressions once during the filter initialization or
10747	when a command is processed
10748
10749	@item frame
10750	evaluate expressions for each incoming frame
10751	@end table
10752
10753	Default value is @samp{init}.
10754	@end table
10755
10756	@section phase
10757
10758	Delay interlaced video by one field time so that the field order changes.
10759
10760	The intended use is to fix PAL movies that have been captured with the
10761	opposite field order to the film-to-video transfer.
10762
10763	A description of the accepted parameters follows.
10764
10765	@table @option
10766	@item mode
10767	Set phase mode.
10768
10769	It accepts the following values:
10770	@table @samp
10771	@item t
10772	Capture field order top-first, transfer bottom-first.
10773	Filter will delay the bottom field.
10774
10775	@item b
10776	Capture field order bottom-first, transfer top-first.
10777	Filter will delay the top field.
10778
10779	@item p
10780	Capture and transfer with the same field order. This mode only exists
10781	for the documentation of the other options to refer to, but if you
10782	actually select it, the filter will faithfully do nothing.
10783
10784	@item a
10785	Capture field order determined automatically by field flags, transfer
10786	opposite.
10787	Filter selects among @samp{t} and @samp{b} modes on a frame by frame
10788	basis using field flags. If no field information is available,
10789	then this works just like @samp{u}.
10790
10791	@item u
10792	Capture unknown or varying, transfer opposite.
10793	Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
10794	analyzing the images and selecting the alternative that produces best
10795	match between the fields.
10796
10797	@item T
10798	Capture top-first, transfer unknown or varying.
10799	Filter selects among @samp{t} and @samp{p} using image analysis.
10800
10801	@item B
10802	Capture bottom-first, transfer unknown or varying.
10803	Filter selects among @samp{b} and @samp{p} using image analysis.
10804
10805	@item A
10806	Capture determined by field flags, transfer unknown or varying.
10807	Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
10808	image analysis. If no field information is available, then this works just
10809	like @samp{U}. This is the default mode.
10810
10811	@item U
10812	Both capture and transfer unknown or varying.
10813	Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
10814	@end table
10815	@end table
10816
10817	@section pixdesctest
10818
10819	Pixel format descriptor test filter, mainly useful for internal
10820	testing. The output video should be equal to the input video.
10821
10822	For example:
10823	@example
10824	format=monow, pixdesctest
10825	@end example
10826
10827	can be used to test the monowhite pixel format descriptor definition.
10828
10829	@section pp
10830
10831	Enable the specified chain of postprocessing subfilters using libpostproc. This
10832	library should be automatically selected with a GPL build (@code{--enable-gpl}).
10833	Subfilters must be separated by '/' and can be disabled by prepending a '-'.
10834	Each subfilter and some options have a short and a long name that can be used
10835	interchangeably, i.e. dr/dering are the same.
10836
10837	The filters accept the following options:
10838
10839	@table @option
10840	@item subfilters
10841	Set postprocessing subfilters string.
10842	@end table
10843
10844	All subfilters share common options to determine their scope:
10845
10846	@table @option
10847	@item a/autoq
10848	Honor the quality commands for this subfilter.
10849
10850	@item c/chrom
10851	Do chrominance filtering, too (default).
10852
10853	@item y/nochrom
10854	Do luminance filtering only (no chrominance).
10855
10856	@item n/noluma
10857	Do chrominance filtering only (no luminance).
10858	@end table
10859
10860	These options can be appended after the subfilter name, separated by a '|'.
10861
10862	Available subfilters are:
10863
10864	@table @option
10865	@item hb/hdeblock[|difference[|flatness]]
10866	Horizontal deblocking filter
10867	@table @option
10868	@item difference
10869	Difference factor where higher values mean more deblocking (default: @code{32}).
10870	@item flatness
10871	Flatness threshold where lower values mean more deblocking (default: @code{39}).
10872	@end table
10873
10874	@item vb/vdeblock[|difference[|flatness]]
10875	Vertical deblocking filter
10876	@table @option
10877	@item difference
10878	Difference factor where higher values mean more deblocking (default: @code{32}).
10879	@item flatness
10880	Flatness threshold where lower values mean more deblocking (default: @code{39}).
10881	@end table
10882
10883	@item ha/hadeblock[|difference[|flatness]]
10884	Accurate horizontal deblocking filter
10885	@table @option
10886	@item difference
10887	Difference factor where higher values mean more deblocking (default: @code{32}).
10888	@item flatness
10889	Flatness threshold where lower values mean more deblocking (default: @code{39}).
10890	@end table
10891
10892	@item va/vadeblock[|difference[|flatness]]
10893	Accurate vertical deblocking filter
10894	@table @option
10895	@item difference
10896	Difference factor where higher values mean more deblocking (default: @code{32}).
10897	@item flatness
10898	Flatness threshold where lower values mean more deblocking (default: @code{39}).
10899	@end table
10900	@end table
10901
10902	The horizontal and vertical deblocking filters share the difference and
10903	flatness values so you cannot set different horizontal and vertical
10904	thresholds.
10905
10906	@table @option
10907	@item h1/x1hdeblock
10908	Experimental horizontal deblocking filter
10909
10910	@item v1/x1vdeblock
10911	Experimental vertical deblocking filter
10912
10913	@item dr/dering
10914	Deringing filter
10915
10916	@item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
10917	@table @option
10918	@item threshold1
10919	larger -> stronger filtering
10920	@item threshold2
10921	larger -> stronger filtering
10922	@item threshold3
10923	larger -> stronger filtering
10924	@end table
10925
10926	@item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
10927	@table @option
10928	@item f/fullyrange
10929	Stretch luminance to @code{0-255}.
10930	@end table
10931
10932	@item lb/linblenddeint
10933	Linear blend deinterlacing filter that deinterlaces the given block by
10934	filtering all lines with a @code{(1 2 1)} filter.
10935
10936	@item li/linipoldeint
10937	Linear interpolating deinterlacing filter that deinterlaces the given block by
10938	linearly interpolating every second line.
10939
10940	@item ci/cubicipoldeint
10941	Cubic interpolating deinterlacing filter deinterlaces the given block by
10942	cubically interpolating every second line.
10943
10944	@item md/mediandeint
10945	Median deinterlacing filter that deinterlaces the given block by applying a
10946	median filter to every second line.
10947
10948	@item fd/ffmpegdeint
10949	FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
10950	second line with a @code{(-1 4 2 4 -1)} filter.
10951
10952	@item l5/lowpass5
10953	Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
10954	block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
10955
10956	@item fq/forceQuant[|quantizer]
10957	Overrides the quantizer table from the input with the constant quantizer you
10958	specify.
10959	@table @option
10960	@item quantizer
10961	Quantizer to use
10962	@end table
10963
10964	@item de/default
10965	Default pp filter combination (@code{hb|a,vb|a,dr|a})
10966
10967	@item fa/fast
10968	Fast pp filter combination (@code{h1|a,v1|a,dr|a})
10969
10970	@item ac
10971	High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
10972	@end table
10973
10974	@subsection Examples
10975
10976	@itemize
10977	@item
10978	Apply horizontal and vertical deblocking, deringing and automatic
10979	brightness/contrast:
10980	@example
10981	pp=hb/vb/dr/al
10982	@end example
10983
10984	@item
10985	Apply default filters without brightness/contrast correction:
10986	@example
10987	pp=de/-al
10988	@end example
10989
10990	@item
10991	Apply default filters and temporal denoiser:
10992	@example
10993	pp=default/tmpnoise|1|2|3
10994	@end example
10995
10996	@item
10997	Apply deblocking on luminance only, and switch vertical deblocking on or off
10998	automatically depending on available CPU time:
10999	@example
11000	pp=hb|y/vb|a
11001	@end example
11002	@end itemize
11003
11004	@section pp7
11005	Apply Postprocessing filter 7. It is variant of the @ref{spp} filter,
11006	similar to spp = 6 with 7 point DCT, where only the center sample is
11007	used after IDCT.
11008
11009	The filter accepts the following options:
11010
11011	@table @option
11012	@item qp
11013	Force a constant quantization parameter. It accepts an integer in range
11014	0 to 63. If not set, the filter will use the QP from the video stream
11015	(if available).
11016
11017	@item mode
11018	Set thresholding mode. Available modes are:
11019
11020	@table @samp
11021	@item hard
11022	Set hard thresholding.
11023	@item soft
11024	Set soft thresholding (better de-ringing effect, but likely blurrier).
11025	@item medium
11026	Set medium thresholding (good results, default).
11027	@end table
11028	@end table
11029
11030	@section premultiply
11031	Apply alpha premultiply effect to input video stream using first plane
11032	of second stream as alpha.
11033
11034	Both streams must have same dimensions and same pixel format.
11035
11036	@section prewitt
11037	Apply prewitt operator to input video stream.
11038
11039	The filter accepts the following option:
11040
11041	@table @option
11042	@item planes
11043	Set which planes will be processed, unprocessed planes will be copied.
11044	By default value 0xf, all planes will be processed.
11045
11046	@item scale
11047	Set value which will be multiplied with filtered result.
11048
11049	@item delta
11050	Set value which will be added to filtered result.
11051	@end table
11052
11053	@section psnr
11054
11055	Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
11056	Ratio) between two input videos.
11057
11058	This filter takes in input two input videos, the first input is
11059	considered the "main" source and is passed unchanged to the
11060	output. The second input is used as a "reference" video for computing
11061	the PSNR.
11062
11063	Both video inputs must have the same resolution and pixel format for
11064	this filter to work correctly. Also it assumes that both inputs
11065	have the same number of frames, which are compared one by one.
11066
11067	The obtained average PSNR is printed through the logging system.
11068
11069	The filter stores the accumulated MSE (mean squared error) of each
11070	frame, and at the end of the processing it is averaged across all frames
11071	equally, and the following formula is applied to obtain the PSNR:
11072
11073	@example
11074	PSNR = 10*log10(MAX^2/MSE)
11075	@end example
11076
11077	Where MAX is the average of the maximum values of each component of the
11078	image.
11079
11080	The description of the accepted parameters follows.
11081
11082	@table @option
11083	@item stats_file, f
11084	If specified the filter will use the named file to save the PSNR of
11085	each individual frame. When filename equals "-" the data is sent to
11086	standard output.
11087
11088	@item stats_version
11089	Specifies which version of the stats file format to use. Details of
11090	each format are written below.
11091	Default value is 1.
11092
11093	@item stats_add_max
11094	Determines whether the max value is output to the stats log.
11095	Default value is 0.
11096	Requires stats_version >= 2. If this is set and stats_version < 2,
11097	the filter will return an error.
11098	@end table
11099
11100	The file printed if @var{stats_file} is selected, contains a sequence of
11101	key/value pairs of the form @var{key}:@var{value} for each compared
11102	couple of frames.
11103
11104	If a @var{stats_version} greater than 1 is specified, a header line precedes
11105	the list of per-frame-pair stats, with key value pairs following the frame
11106	format with the following parameters:
11107
11108	@table @option
11109	@item psnr_log_version
11110	The version of the log file format. Will match @var{stats_version}.
11111
11112	@item fields
11113	A comma separated list of the per-frame-pair parameters included in
11114	the log.
11115	@end table
11116
11117	A description of each shown per-frame-pair parameter follows:
11118
11119	@table @option
11120	@item n
11121	sequential number of the input frame, starting from 1
11122
11123	@item mse_avg
11124	Mean Square Error pixel-by-pixel average difference of the compared
11125	frames, averaged over all the image components.
11126
11127	@item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
11128	Mean Square Error pixel-by-pixel average difference of the compared
11129	frames for the component specified by the suffix.
11130
11131	@item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
11132	Peak Signal to Noise ratio of the compared frames for the component
11133	specified by the suffix.
11134
11135	@item max_avg, max_y, max_u, max_v
11136	Maximum allowed value for each channel, and average over all
11137	channels.
11138	@end table
11139
11140	For example:
11141	@example
11142	movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
11143	[main][ref] psnr="stats_file=stats.log" [out]
11144	@end example
11145
11146	On this example the input file being processed is compared with the
11147	reference file @file{ref_movie.mpg}. The PSNR of each individual frame
11148	is stored in @file{stats.log}.
11149
11150	@anchor{pullup}
11151	@section pullup
11152
11153	Pulldown reversal (inverse telecine) filter, capable of handling mixed
11154	hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
11155	content.
11156
11157	The pullup filter is designed to take advantage of future context in making
11158	its decisions. This filter is stateless in the sense that it does not lock
11159	onto a pattern to follow, but it instead looks forward to the following
11160	fields in order to identify matches and rebuild progressive frames.
11161
11162	To produce content with an even framerate, insert the fps filter after
11163	pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
11164	@code{fps=24} for 30fps and the (rare) telecined 25fps input.
11165
11166	The filter accepts the following options:
11167
11168	@table @option
11169	@item jl
11170	@item jr
11171	@item jt
11172	@item jb
11173	These options set the amount of "junk" to ignore at the left, right, top, and
11174	bottom of the image, respectively. Left and right are in units of 8 pixels,
11175	while top and bottom are in units of 2 lines.
11176	The default is 8 pixels on each side.
11177
11178	@item sb
11179	Set the strict breaks. Setting this option to 1 will reduce the chances of
11180	filter generating an occasional mismatched frame, but it may also cause an
11181	excessive number of frames to be dropped during high motion sequences.
11182	Conversely, setting it to -1 will make filter match fields more easily.
11183	This may help processing of video where there is slight blurring between
11184	the fields, but may also cause there to be interlaced frames in the output.
11185	Default value is @code{0}.
11186
11187	@item mp
11188	Set the metric plane to use. It accepts the following values:
11189	@table @samp
11190	@item l
11191	Use luma plane.
11192
11193	@item u
11194	Use chroma blue plane.
11195
11196	@item v
11197	Use chroma red plane.
11198	@end table
11199
11200	This option may be set to use chroma plane instead of the default luma plane
11201	for doing filter's computations. This may improve accuracy on very clean
11202	source material, but more likely will decrease accuracy, especially if there
11203	is chroma noise (rainbow effect) or any grayscale video.
11204	The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
11205	load and make pullup usable in realtime on slow machines.
11206	@end table
11207
11208	For best results (without duplicated frames in the output file) it is
11209	necessary to change the output frame rate. For example, to inverse
11210	telecine NTSC input:
11211	@example
11212	ffmpeg -i input -vf pullup -r 24000/1001 ...
11213	@end example
11214
11215	@section qp
11216
11217	Change video quantization parameters (QP).
11218
11219	The filter accepts the following option:
11220
11221	@table @option
11222	@item qp
11223	Set expression for quantization parameter.
11224	@end table
11225
11226	The expression is evaluated through the eval API and can contain, among others,
11227	the following constants:
11228
11229	@table @var
11230	@item known
11231	1 if index is not 129, 0 otherwise.
11232
11233	@item qp
11234	Sequentional index starting from -129 to 128.
11235	@end table
11236
11237	@subsection Examples
11238
11239	@itemize
11240	@item
11241	Some equation like:
11242	@example
11243	qp=2+2*sin(PI*qp)
11244	@end example
11245	@end itemize
11246
11247	@section random
11248
11249	Flush video frames from internal cache of frames into a random order.
11250	No frame is discarded.
11251	Inspired by @ref{frei0r} nervous filter.
11252
11253	@table @option
11254	@item frames
11255	Set size in number of frames of internal cache, in range from @code{2} to
11256	@code{512}. Default is @code{30}.
11257
11258	@item seed
11259	Set seed for random number generator, must be an integer included between
11260	@code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
11261	less than @code{0}, the filter will try to use a good random seed on a
11262	best effort basis.
11263	@end table
11264
11265	@section readeia608
11266
11267	Read closed captioning (EIA-608) information from the top lines of a video frame.
11268
11269	This filter adds frame metadata for @code{lavfi.readeia608.X.cc} and
11270	@code{lavfi.readeia608.X.line}, where @code{X} is the number of the identified line
11271	with EIA-608 data (starting from 0). A description of each metadata value follows:
11272
11273	@table @option
11274	@item lavfi.readeia608.X.cc
11275	The two bytes stored as EIA-608 data (printed in hexadecimal).
11276
11277	@item lavfi.readeia608.X.line
11278	The number of the line on which the EIA-608 data was identified and read.
11279	@end table
11280
11281	This filter accepts the following options:
11282
11283	@table @option
11284	@item scan_min
11285	Set the line to start scanning for EIA-608 data. Default is @code{0}.
11286
11287	@item scan_max
11288	Set the line to end scanning for EIA-608 data. Default is @code{29}.
11289
11290	@item mac
11291	Set minimal acceptable amplitude change for sync codes detection.
11292	Default is @code{0.2}. Allowed range is @code{[0.001 - 1]}.
11293
11294	@item spw
11295	Set the ratio of width reserved for sync code detection.
11296	Default is @code{0.27}. Allowed range is @code{[0.01 - 0.7]}.
11297
11298	@item mhd
11299	Set the max peaks height difference for sync code detection.
11300	Default is @code{0.1}. Allowed range is @code{[0.0 - 0.5]}.
11301
11302	@item mpd
11303	Set max peaks period difference for sync code detection.
11304	Default is @code{0.1}. Allowed range is @code{[0.0 - 0.5]}.
11305
11306	@item msd
11307	Set the first two max start code bits differences.
11308	Default is @code{0.02}. Allowed range is @code{[0.0 - 0.5]}.
11309
11310	@item bhd
11311	Set the minimum ratio of bits height compared to 3rd start code bit.
11312	Default is @code{0.75}. Allowed range is @code{[0.01 - 1]}.
11313
11314	@item th_w
11315	Set the white color threshold. Default is @code{0.35}. Allowed range is @code{[0.1 - 1]}.
11316
11317	@item th_b
11318	Set the black color threshold. Default is @code{0.15}. Allowed range is @code{[0.0 - 0.5]}.
11319
11320	@item chp
11321	Enable checking the parity bit. In the event of a parity error, the filter will output
11322	@code{0x00} for that character. Default is false.
11323	@end table
11324
11325	@subsection Examples
11326
11327	@itemize
11328	@item
11329	Output a csv with presentation time and the first two lines of identified EIA-608 captioning data.
11330	@example
11331	ffprobe -f lavfi -i movie=captioned_video.mov,readeia608 -show_entries frame=pkt_pts_time:frame_tags=lavfi.readeia608.0.cc,lavfi.readeia608.1.cc -of csv
11332	@end example
11333	@end itemize
11334
11335	@section readvitc
11336
11337	Read vertical interval timecode (VITC) information from the top lines of a
11338	video frame.
11339
11340	The filter adds frame metadata key @code{lavfi.readvitc.tc_str} with the
11341	timecode value, if a valid timecode has been detected. Further metadata key
11342	@code{lavfi.readvitc.found} is set to 0/1 depending on whether
11343	timecode data has been found or not.
11344
11345	This filter accepts the following options:
11346
11347	@table @option
11348	@item scan_max
11349	Set the maximum number of lines to scan for VITC data. If the value is set to
11350	@code{-1} the full video frame is scanned. Default is @code{45}.
11351
11352	@item thr_b
11353	Set the luma threshold for black. Accepts float numbers in the range [0.0,1.0],
11354	default value is @code{0.2}. The value must be equal or less than @code{thr_w}.
11355
11356	@item thr_w
11357	Set the luma threshold for white. Accepts float numbers in the range [0.0,1.0],
11358	default value is @code{0.6}. The value must be equal or greater than @code{thr_b}.
11359	@end table
11360
11361	@subsection Examples
11362
11363	@itemize
11364	@item
11365	Detect and draw VITC data onto the video frame; if no valid VITC is detected,
11366	draw @code{--:--:--:--} as a placeholder:
11367	@example
11368	ffmpeg -i input.avi -filter:v 'readvitc,drawtext=fontfile=FreeMono.ttf:text=%@{metadata\\:lavfi.readvitc.tc_str\\:--\\\\\\:--\\\\\\:--\\\\\\:--@}:x=(w-tw)/2:y=400-ascent'
11369	@end example
11370	@end itemize
11371
11372	@section remap
11373
11374	Remap pixels using 2nd: Xmap and 3rd: Ymap input video stream.
11375
11376	Destination pixel at position (X, Y) will be picked from source (x, y) position
11377	where x = Xmap(X, Y) and y = Ymap(X, Y). If mapping values are out of range, zero
11378	value for pixel will be used for destination pixel.
11379
11380	Xmap and Ymap input video streams must be of same dimensions. Output video stream
11381	will have Xmap/Ymap video stream dimensions.
11382	Xmap and Ymap input video streams are 16bit depth, single channel.
11383
11384	@section removegrain
11385
11386	The removegrain filter is a spatial denoiser for progressive video.
11387
11388	@table @option
11389	@item m0
11390	Set mode for the first plane.
11391
11392	@item m1
11393	Set mode for the second plane.
11394
11395	@item m2
11396	Set mode for the third plane.
11397
11398	@item m3
11399	Set mode for the fourth plane.
11400	@end table
11401
11402	Range of mode is from 0 to 24. Description of each mode follows:
11403
11404	@table @var
11405	@item 0
11406	Leave input plane unchanged. Default.
11407
11408	@item 1
11409	Clips the pixel with the minimum and maximum of the 8 neighbour pixels.
11410
11411	@item 2
11412	Clips the pixel with the second minimum and maximum of the 8 neighbour pixels.
11413
11414	@item 3
11415	Clips the pixel with the third minimum and maximum of the 8 neighbour pixels.
11416
11417	@item 4
11418	Clips the pixel with the fourth minimum and maximum of the 8 neighbour pixels.
11419	This is equivalent to a median filter.
11420
11421	@item 5
11422	Line-sensitive clipping giving the minimal change.
11423
11424	@item 6
11425	Line-sensitive clipping, intermediate.
11426
11427	@item 7
11428	Line-sensitive clipping, intermediate.
11429
11430	@item 8
11431	Line-sensitive clipping, intermediate.
11432
11433	@item 9
11434	Line-sensitive clipping on a line where the neighbours pixels are the closest.
11435
11436	@item 10
11437	Replaces the target pixel with the closest neighbour.
11438
11439	@item 11
11440	[1 2 1] horizontal and vertical kernel blur.
11441
11442	@item 12
11443	Same as mode 11.
11444
11445	@item 13
11446	Bob mode, interpolates top field from the line where the neighbours
11447	pixels are the closest.
11448
11449	@item 14
11450	Bob mode, interpolates bottom field from the line where the neighbours
11451	pixels are the closest.
11452
11453	@item 15
11454	Bob mode, interpolates top field. Same as 13 but with a more complicated
11455	interpolation formula.
11456
11457	@item 16
11458	Bob mode, interpolates bottom field. Same as 14 but with a more complicated
11459	interpolation formula.
11460
11461	@item 17
11462	Clips the pixel with the minimum and maximum of respectively the maximum and
11463	minimum of each pair of opposite neighbour pixels.
11464
11465	@item 18
11466	Line-sensitive clipping using opposite neighbours whose greatest distance from
11467	the current pixel is minimal.
11468
11469	@item 19
11470	Replaces the pixel with the average of its 8 neighbours.
11471
11472	@item 20
11473	Averages the 9 pixels ([1 1 1] horizontal and vertical blur).
11474
11475	@item 21
11476	Clips pixels using the averages of opposite neighbour.
11477
11478	@item 22
11479	Same as mode 21 but simpler and faster.
11480
11481	@item 23
11482	Small edge and halo removal, but reputed useless.
11483
11484	@item 24
11485	Similar as 23.
11486	@end table
11487
11488	@section removelogo
11489
11490	Suppress a TV station logo, using an image file to determine which
11491	pixels comprise the logo. It works by filling in the pixels that
11492	comprise the logo with neighboring pixels.
11493
11494	The filter accepts the following options:
11495
11496	@table @option
11497	@item filename, f
11498	Set the filter bitmap file, which can be any image format supported by
11499	libavformat. The width and height of the image file must match those of the
11500	video stream being processed.
11501	@end table
11502
11503	Pixels in the provided bitmap image with a value of zero are not
11504	considered part of the logo, non-zero pixels are considered part of
11505	the logo. If you use white (255) for the logo and black (0) for the
11506	rest, you will be safe. For making the filter bitmap, it is
11507	recommended to take a screen capture of a black frame with the logo
11508	visible, and then using a threshold filter followed by the erode
11509	filter once or twice.
11510
11511	If needed, little splotches can be fixed manually. Remember that if
11512	logo pixels are not covered, the filter quality will be much
11513	reduced. Marking too many pixels as part of the logo does not hurt as
11514	much, but it will increase the amount of blurring needed to cover over
11515	the image and will destroy more information than necessary, and extra
11516	pixels will slow things down on a large logo.
11517
11518	@section repeatfields
11519
11520	This filter uses the repeat_field flag from the Video ES headers and hard repeats
11521	fields based on its value.
11522
11523	@section reverse
11524
11525	Reverse a video clip.
11526
11527	Warning: This filter requires memory to buffer the entire clip, so trimming
11528	is suggested.
11529
11530	@subsection Examples
11531
11532	@itemize
11533	@item
11534	Take the first 5 seconds of a clip, and reverse it.
11535	@example
11536	trim=end=5,reverse
11537	@end example
11538	@end itemize
11539
11540	@section rotate
11541
11542	Rotate video by an arbitrary angle expressed in radians.
11543
11544	The filter accepts the following options:
11545
11546	A description of the optional parameters follows.
11547	@table @option
11548	@item angle, a
11549	Set an expression for the angle by which to rotate the input video
11550	clockwise, expressed as a number of radians. A negative value will
11551	result in a counter-clockwise rotation. By default it is set to "0".
11552
11553	This expression is evaluated for each frame.
11554
11555	@item out_w, ow
11556	Set the output width expression, default value is "iw".
11557	This expression is evaluated just once during configuration.
11558
11559	@item out_h, oh
11560	Set the output height expression, default value is "ih".
11561	This expression is evaluated just once during configuration.
11562
11563	@item bilinear
11564	Enable bilinear interpolation if set to 1, a value of 0 disables
11565	it. Default value is 1.
11566
11567	@item fillcolor, c
11568	Set the color used to fill the output area not covered by the rotated
11569	image. For the general syntax of this option, check the "Color" section in the
11570	ffmpeg-utils manual. If the special value "none" is selected then no
11571	background is printed (useful for example if the background is never shown).
11572
11573	Default value is "black".
11574	@end table
11575
11576	The expressions for the angle and the output size can contain the
11577	following constants and functions:
11578
11579	@table @option
11580	@item n
11581	sequential number of the input frame, starting from 0. It is always NAN
11582	before the first frame is filtered.
11583
11584	@item t
11585	time in seconds of the input frame, it is set to 0 when the filter is
11586	configured. It is always NAN before the first frame is filtered.
11587
11588	@item hsub
11589	@item vsub
11590	horizontal and vertical chroma subsample values. For example for the
11591	pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11592
11593	@item in_w, iw
11594	@item in_h, ih
11595	the input video width and height
11596
11597	@item out_w, ow
11598	@item out_h, oh
11599	the output width and height, that is the size of the padded area as
11600	specified by the @var{width} and @var{height} expressions
11601
11602	@item rotw(a)
11603	@item roth(a)
11604	the minimal width/height required for completely containing the input
11605	video rotated by @var{a} radians.
11606
11607	These are only available when computing the @option{out_w} and
11608	@option{out_h} expressions.
11609	@end table
11610
11611	@subsection Examples
11612
11613	@itemize
11614	@item
11615	Rotate the input by PI/6 radians clockwise:
11616	@example
11617	rotate=PI/6
11618	@end example
11619
11620	@item
11621	Rotate the input by PI/6 radians counter-clockwise:
11622	@example
11623	rotate=-PI/6
11624	@end example
11625
11626	@item
11627	Rotate the input by 45 degrees clockwise:
11628	@example
11629	rotate=45*PI/180
11630	@end example
11631
11632	@item
11633	Apply a constant rotation with period T, starting from an angle of PI/3:
11634	@example
11635	rotate=PI/3+2*PI*t/T
11636	@end example
11637
11638	@item
11639	Make the input video rotation oscillating with a period of T
11640	seconds and an amplitude of A radians:
11641	@example
11642	rotate=A*sin(2*PI/T*t)
11643	@end example
11644
11645	@item
11646	Rotate the video, output size is chosen so that the whole rotating
11647	input video is always completely contained in the output:
11648	@example
11649	rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
11650	@end example
11651
11652	@item
11653	Rotate the video, reduce the output size so that no background is ever
11654	shown:
11655	@example
11656	rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
11657	@end example
11658	@end itemize
11659
11660	@subsection Commands
11661
11662	The filter supports the following commands:
11663
11664	@table @option
11665	@item a, angle
11666	Set the angle expression.
11667	The command accepts the same syntax of the corresponding option.
11668
11669	If the specified expression is not valid, it is kept at its current
11670	value.
11671	@end table
11672
11673	@section sab
11674
11675	Apply Shape Adaptive Blur.
11676
11677	The filter accepts the following options:
11678
11679	@table @option
11680	@item luma_radius, lr
11681	Set luma blur filter strength, must be a value in range 0.1-4.0, default
11682	value is 1.0. A greater value will result in a more blurred image, and
11683	in slower processing.
11684
11685	@item luma_pre_filter_radius, lpfr
11686	Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
11687	value is 1.0.
11688
11689	@item luma_strength, ls
11690	Set luma maximum difference between pixels to still be considered, must
11691	be a value in the 0.1-100.0 range, default value is 1.0.
11692
11693	@item chroma_radius, cr
11694	Set chroma blur filter strength, must be a value in range -0.9-4.0. A
11695	greater value will result in a more blurred image, and in slower
11696	processing.
11697
11698	@item chroma_pre_filter_radius, cpfr
11699	Set chroma pre-filter radius, must be a value in the -0.9-2.0 range.
11700
11701	@item chroma_strength, cs
11702	Set chroma maximum difference between pixels to still be considered,
11703	must be a value in the -0.9-100.0 range.
11704	@end table
11705
11706	Each chroma option value, if not explicitly specified, is set to the
11707	corresponding luma option value.
11708
11709	@anchor{scale}
11710	@section scale
11711
11712	Scale (resize) the input video, using the libswscale library.
11713
11714	The scale filter forces the output display aspect ratio to be the same
11715	of the input, by changing the output sample aspect ratio.
11716
11717	If the input image format is different from the format requested by
11718	the next filter, the scale filter will convert the input to the
11719	requested format.
11720
11721	@subsection Options
11722	The filter accepts the following options, or any of the options
11723	supported by the libswscale scaler.
11724
11725	See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
11726	the complete list of scaler options.
11727
11728	@table @option
11729	@item width, w
11730	@item height, h
11731	Set the output video dimension expression. Default value is the input
11732	dimension.
11733
11734	If the value is 0, the input width is used for the output.
11735
11736	If one of the values is -1, the scale filter will use a value that
11737	maintains the aspect ratio of the input image, calculated from the
11738	other specified dimension. If both of them are -1, the input size is
11739	used
11740
11741	If one of the values is -n with n > 1, the scale filter will also use a value
11742	that maintains the aspect ratio of the input image, calculated from the other
11743	specified dimension. After that it will, however, make sure that the calculated
11744	dimension is divisible by n and adjust the value if necessary.
11745
11746	See below for the list of accepted constants for use in the dimension
11747	expression.
11748
11749	@item eval
11750	Specify when to evaluate @var{width} and @var{height} expression. It accepts the following values:
11751
11752	@table @samp
11753	@item init
11754	Only evaluate expressions once during the filter initialization or when a command is processed.
11755
11756	@item frame
11757	Evaluate expressions for each incoming frame.
11758
11759	@end table
11760
11761	Default value is @samp{init}.
11762
11763
11764	@item interl
11765	Set the interlacing mode. It accepts the following values:
11766
11767	@table @samp
11768	@item 1
11769	Force interlaced aware scaling.
11770
11771	@item 0
11772	Do not apply interlaced scaling.
11773
11774	@item -1
11775	Select interlaced aware scaling depending on whether the source frames
11776	are flagged as interlaced or not.
11777	@end table
11778
11779	Default value is @samp{0}.
11780
11781	@item flags
11782	Set libswscale scaling flags. See
11783	@ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11784	complete list of values. If not explicitly specified the filter applies
11785	the default flags.
11786
11787
11788	@item param0, param1
11789	Set libswscale input parameters for scaling algorithms that need them. See
11790	@ref{sws_params,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11791	complete documentation. If not explicitly specified the filter applies
11792	empty parameters.
11793
11794
11795
11796	@item size, s
11797	Set the video size. For the syntax of this option, check the
11798	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11799
11800	@item in_color_matrix
11801	@item out_color_matrix
11802	Set in/output YCbCr color space type.
11803
11804	This allows the autodetected value to be overridden as well as allows forcing
11805	a specific value used for the output and encoder.
11806
11807	If not specified, the color space type depends on the pixel format.
11808
11809	Possible values:
11810
11811	@table @samp
11812	@item auto
11813	Choose automatically.
11814
11815	@item bt709
11816	Format conforming to International Telecommunication Union (ITU)
11817	Recommendation BT.709.
11818
11819	@item fcc
11820	Set color space conforming to the United States Federal Communications
11821	Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
11822
11823	@item bt601
11824	Set color space conforming to:
11825
11826	@itemize
11827	@item
11828	ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
11829
11830	@item
11831	ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
11832
11833	@item
11834	Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
11835
11836	@end itemize
11837
11838	@item smpte240m
11839	Set color space conforming to SMPTE ST 240:1999.
11840	@end table
11841
11842	@item in_range
11843	@item out_range
11844	Set in/output YCbCr sample range.
11845
11846	This allows the autodetected value to be overridden as well as allows forcing
11847	a specific value used for the output and encoder. If not specified, the
11848	range depends on the pixel format. Possible values:
11849
11850	@table @samp
11851	@item auto
11852	Choose automatically.
11853
11854	@item jpeg/full/pc
11855	Set full range (0-255 in case of 8-bit luma).
11856
11857	@item mpeg/tv
11858	Set "MPEG" range (16-235 in case of 8-bit luma).
11859	@end table
11860
11861	@item force_original_aspect_ratio
11862	Enable decreasing or increasing output video width or height if necessary to
11863	keep the original aspect ratio. Possible values:
11864
11865	@table @samp
11866	@item disable
11867	Scale the video as specified and disable this feature.
11868
11869	@item decrease
11870	The output video dimensions will automatically be decreased if needed.
11871
11872	@item increase
11873	The output video dimensions will automatically be increased if needed.
11874
11875	@end table
11876
11877	One useful instance of this option is that when you know a specific device's
11878	maximum allowed resolution, you can use this to limit the output video to
11879	that, while retaining the aspect ratio. For example, device A allows
11880	1280x720 playback, and your video is 1920x800. Using this option (set it to
11881	decrease) and specifying 1280x720 to the command line makes the output
11882	1280x533.
11883
11884	Please note that this is a different thing than specifying -1 for @option{w}
11885	or @option{h}, you still need to specify the output resolution for this option
11886	to work.
11887
11888	@end table
11889
11890	The values of the @option{w} and @option{h} options are expressions
11891	containing the following constants:
11892
11893	@table @var
11894	@item in_w
11895	@item in_h
11896	The input width and height
11897
11898	@item iw
11899	@item ih
11900	These are the same as @var{in_w} and @var{in_h}.
11901
11902	@item out_w
11903	@item out_h
11904	The output (scaled) width and height
11905
11906	@item ow
11907	@item oh
11908	These are the same as @var{out_w} and @var{out_h}
11909
11910	@item a
11911	The same as @var{iw} / @var{ih}
11912
11913	@item sar
11914	input sample aspect ratio
11915
11916	@item dar
11917	The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
11918
11919	@item hsub
11920	@item vsub
11921	horizontal and vertical input chroma subsample values. For example for the
11922	pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11923
11924	@item ohsub
11925	@item ovsub
11926	horizontal and vertical output chroma subsample values. For example for the
11927	pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11928	@end table
11929
11930	@subsection Examples
11931
11932	@itemize
11933	@item
11934	Scale the input video to a size of 200x100
11935	@example
11936	scale=w=200:h=100
11937	@end example
11938
11939	This is equivalent to:
11940	@example
11941	scale=200:100
11942	@end example
11943
11944	or:
11945	@example
11946	scale=200x100
11947	@end example
11948
11949	@item
11950	Specify a size abbreviation for the output size:
11951	@example
11952	scale=qcif
11953	@end example
11954
11955	which can also be written as:
11956	@example
11957	scale=size=qcif
11958	@end example
11959
11960	@item
11961	Scale the input to 2x:
11962	@example
11963	scale=w=2*iw:h=2*ih
11964	@end example
11965
11966	@item
11967	The above is the same as:
11968	@example
11969	scale=2*in_w:2*in_h
11970	@end example
11971
11972	@item
11973	Scale the input to 2x with forced interlaced scaling:
11974	@example
11975	scale=2*iw:2*ih:interl=1
11976	@end example
11977
11978	@item
11979	Scale the input to half size:
11980	@example
11981	scale=w=iw/2:h=ih/2
11982	@end example
11983
11984	@item
11985	Increase the width, and set the height to the same size:
11986	@example
11987	scale=3/2*iw:ow
11988	@end example
11989
11990	@item
11991	Seek Greek harmony:
11992	@example
11993	scale=iw:1/PHI*iw
11994	scale=ih*PHI:ih
11995	@end example
11996
11997	@item
11998	Increase the height, and set the width to 3/2 of the height:
11999	@example
12000	scale=w=3/2*oh:h=3/5*ih
12001	@end example
12002
12003	@item
12004	Increase the size, making the size a multiple of the chroma
12005	subsample values:
12006	@example
12007	scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
12008	@end example
12009
12010	@item
12011	Increase the width to a maximum of 500 pixels,
12012	keeping the same aspect ratio as the input:
12013	@example
12014	scale=w='min(500\, iw*3/2):h=-1'
12015	@end example
12016	@end itemize
12017
12018	@subsection Commands
12019
12020	This filter supports the following commands:
12021	@table @option
12022	@item width, w
12023	@item height, h
12024	Set the output video dimension expression.
12025	The command accepts the same syntax of the corresponding option.
12026
12027	If the specified expression is not valid, it is kept at its current
12028	value.
12029	@end table
12030
12031	@section scale_npp
12032
12033	Use the NVIDIA Performance Primitives (libnpp) to perform scaling and/or pixel
12034	format conversion on CUDA video frames. Setting the output width and height
12035	works in the same way as for the @var{scale} filter.
12036
12037	The following additional options are accepted:
12038	@table @option
12039	@item format
12040	The pixel format of the output CUDA frames. If set to the string "same" (the
12041	default), the input format will be kept. Note that automatic format negotiation
12042	and conversion is not yet supported for hardware frames
12043
12044	@item interp_algo
12045	The interpolation algorithm used for resizing. One of the following:
12046	@table @option
12047	@item nn
12048	Nearest neighbour.
12049
12050	@item linear
12051	@item cubic
12052	@item cubic2p_bspline
12053	2-parameter cubic (B=1, C=0)
12054
12055	@item cubic2p_catmullrom
12056	2-parameter cubic (B=0, C=1/2)
12057
12058	@item cubic2p_b05c03
12059	2-parameter cubic (B=1/2, C=3/10)
12060
12061	@item super
12062	Supersampling
12063
12064	@item lanczos
12065	@end table
12066
12067	@end table
12068
12069	@section scale2ref
12070
12071	Scale (resize) the input video, based on a reference video.
12072
12073	See the scale filter for available options, scale2ref supports the same but
12074	uses the reference video instead of the main input as basis.
12075
12076	@subsection Examples
12077
12078	@itemize
12079	@item
12080	Scale a subtitle stream to match the main video in size before overlaying
12081	@example
12082	'scale2ref[b][a];[a][b]overlay'
12083	@end example
12084	@end itemize
12085
12086	@anchor{selectivecolor}
12087	@section selectivecolor
12088
12089	Adjust cyan, magenta, yellow and black (CMYK) to certain ranges of colors (such
12090	as "reds", "yellows", "greens", "cyans", ...). The adjustment range is defined
12091	by the "purity" of the color (that is, how saturated it already is).
12092
12093	This filter is similar to the Adobe Photoshop Selective Color tool.
12094
12095	The filter accepts the following options:
12096
12097	@table @option
12098	@item correction_method
12099	Select color correction method.
12100
12101	Available values are:
12102	@table @samp
12103	@item absolute
12104	Specified adjustments are applied "as-is" (added/subtracted to original pixel
12105	component value).
12106	@item relative
12107	Specified adjustments are relative to the original component value.
12108	@end table
12109	Default is @code{absolute}.
12110	@item reds
12111	Adjustments for red pixels (pixels where the red component is the maximum)
12112	@item yellows
12113	Adjustments for yellow pixels (pixels where the blue component is the minimum)
12114	@item greens
12115	Adjustments for green pixels (pixels where the green component is the maximum)
12116	@item cyans
12117	Adjustments for cyan pixels (pixels where the red component is the minimum)
12118	@item blues
12119	Adjustments for blue pixels (pixels where the blue component is the maximum)
12120	@item magentas
12121	Adjustments for magenta pixels (pixels where the green component is the minimum)
12122	@item whites
12123	Adjustments for white pixels (pixels where all components are greater than 128)
12124	@item neutrals
12125	Adjustments for all pixels except pure black and pure white
12126	@item blacks
12127	Adjustments for black pixels (pixels where all components are lesser than 128)
12128	@item psfile
12129	Specify a Photoshop selective color file (@code{.asv}) to import the settings from.
12130	@end table
12131
12132	All the adjustment settings (@option{reds}, @option{yellows}, ...) accept up to
12133	4 space separated floating point adjustment values in the [-1,1] range,
12134	respectively to adjust the amount of cyan, magenta, yellow and black for the
12135	pixels of its range.
12136
12137	@subsection Examples
12138
12139	@itemize
12140	@item
12141	Increase cyan by 50% and reduce yellow by 33% in every green areas, and
12142	increase magenta by 27% in blue areas:
12143	@example
12144	selectivecolor=greens=.5 0 -.33 0:blues=0 .27
12145	@end example
12146
12147	@item
12148	Use a Photoshop selective color preset:
12149	@example
12150	selectivecolor=psfile=MySelectiveColorPresets/Misty.asv
12151	@end example
12152	@end itemize
12153
12154	@anchor{separatefields}
12155	@section separatefields
12156
12157	The @code{separatefields} takes a frame-based video input and splits
12158	each frame into its components fields, producing a new half height clip
12159	with twice the frame rate and twice the frame count.
12160
12161	This filter use field-dominance information in frame to decide which
12162	of each pair of fields to place first in the output.
12163	If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
12164
12165	@section setdar, setsar
12166
12167	The @code{setdar} filter sets the Display Aspect Ratio for the filter
12168	output video.
12169
12170	This is done by changing the specified Sample (aka Pixel) Aspect
12171	Ratio, according to the following equation:
12172	@example
12173	@var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
12174	@end example
12175
12176	Keep in mind that the @code{setdar} filter does not modify the pixel
12177	dimensions of the video frame. Also, the display aspect ratio set by
12178	this filter may be changed by later filters in the filterchain,
12179	e.g. in case of scaling or if another "setdar" or a "setsar" filter is
12180	applied.
12181
12182	The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
12183	the filter output video.
12184
12185	Note that as a consequence of the application of this filter, the
12186	output display aspect ratio will change according to the equation
12187	above.
12188
12189	Keep in mind that the sample aspect ratio set by the @code{setsar}
12190	filter may be changed by later filters in the filterchain, e.g. if
12191	another "setsar" or a "setdar" filter is applied.
12192
12193	It accepts the following parameters:
12194
12195	@table @option
12196	@item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
12197	Set the aspect ratio used by the filter.
12198
12199	The parameter can be a floating point number string, an expression, or
12200	a string of the form @var{num}:@var{den}, where @var{num} and
12201	@var{den} are the numerator and denominator of the aspect ratio. If
12202	the parameter is not specified, it is assumed the value "0".
12203	In case the form "@var{num}:@var{den}" is used, the @code{:} character
12204	should be escaped.
12205
12206	@item max
12207	Set the maximum integer value to use for expressing numerator and
12208	denominator when reducing the expressed aspect ratio to a rational.
12209	Default value is @code{100}.
12210
12211	@end table
12212
12213	The parameter @var{sar} is an expression containing
12214	the following constants:
12215
12216	@table @option
12217	@item E, PI, PHI
12218	These are approximated values for the mathematical constants e
12219	(Euler's number), pi (Greek pi), and phi (the golden ratio).
12220
12221	@item w, h
12222	The input width and height.
12223
12224	@item a
12225	These are the same as @var{w} / @var{h}.
12226
12227	@item sar
12228	The input sample aspect ratio.
12229
12230	@item dar
12231	The input display aspect ratio. It is the same as
12232	(@var{w} / @var{h}) * @var{sar}.
12233
12234	@item hsub, vsub
12235	Horizontal and vertical chroma subsample values. For example, for the
12236	pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12237	@end table
12238
12239	@subsection Examples
12240
12241	@itemize
12242
12243	@item
12244	To change the display aspect ratio to 16:9, specify one of the following:
12245	@example
12246	setdar=dar=1.77777
12247	setdar=dar=16/9
12248	@end example
12249
12250	@item
12251	To change the sample aspect ratio to 10:11, specify:
12252	@example
12253	setsar=sar=10/11
12254	@end example
12255
12256	@item
12257	To set a display aspect ratio of 16:9, and specify a maximum integer value of
12258	1000 in the aspect ratio reduction, use the command:
12259	@example
12260	setdar=ratio=16/9:max=1000
12261	@end example
12262
12263	@end itemize
12264
12265	@anchor{setfield}
12266	@section setfield
12267
12268	Force field for the output video frame.
12269
12270	The @code{setfield} filter marks the interlace type field for the
12271	output frames. It does not change the input frame, but only sets the
12272	corresponding property, which affects how the frame is treated by
12273	following filters (e.g. @code{fieldorder} or @code{yadif}).
12274
12275	The filter accepts the following options:
12276
12277	@table @option
12278
12279	@item mode
12280	Available values are:
12281
12282	@table @samp
12283	@item auto
12284	Keep the same field property.
12285
12286	@item bff
12287	Mark the frame as bottom-field-first.
12288
12289	@item tff
12290	Mark the frame as top-field-first.
12291
12292	@item prog
12293	Mark the frame as progressive.
12294	@end table
12295	@end table
12296
12297	@section showinfo
12298
12299	Show a line containing various information for each input video frame.
12300	The input video is not modified.
12301
12302	The shown line contains a sequence of key/value pairs of the form
12303	@var{key}:@var{value}.
12304
12305	The following values are shown in the output:
12306
12307	@table @option
12308	@item n
12309	The (sequential) number of the input frame, starting from 0.
12310
12311	@item pts
12312	The Presentation TimeStamp of the input frame, expressed as a number of
12313	time base units. The time base unit depends on the filter input pad.
12314
12315	@item pts_time
12316	The Presentation TimeStamp of the input frame, expressed as a number of
12317	seconds.
12318
12319	@item pos
12320	The position of the frame in the input stream, or -1 if this information is
12321	unavailable and/or meaningless (for example in case of synthetic video).
12322
12323	@item fmt
12324	The pixel format name.
12325
12326	@item sar
12327	The sample aspect ratio of the input frame, expressed in the form
12328	@var{num}/@var{den}.
12329
12330	@item s
12331	The size of the input frame. For the syntax of this option, check the
12332	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12333
12334	@item i
12335	The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
12336	for bottom field first).
12337
12338	@item iskey
12339	This is 1 if the frame is a key frame, 0 otherwise.
12340
12341	@item type
12342	The picture type of the input frame ("I" for an I-frame, "P" for a
12343	P-frame, "B" for a B-frame, or "?" for an unknown type).
12344	Also refer to the documentation of the @code{AVPictureType} enum and of
12345	the @code{av_get_picture_type_char} function defined in
12346	@file{libavutil/avutil.h}.
12347
12348	@item checksum
12349	The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
12350
12351	@item plane_checksum
12352	The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
12353	expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
12354	@end table
12355
12356	@section showpalette
12357
12358	Displays the 256 colors palette of each frame. This filter is only relevant for
12359	@var{pal8} pixel format frames.
12360
12361	It accepts the following option:
12362
12363	@table @option
12364	@item s
12365	Set the size of the box used to represent one palette color entry. Default is
12366	@code{30} (for a @code{30x30} pixel box).
12367	@end table
12368
12369	@section shuffleframes
12370
12371	Reorder and/or duplicate and/or drop video frames.
12372
12373	It accepts the following parameters:
12374
12375	@table @option
12376	@item mapping
12377	Set the destination indexes of input frames.
12378	This is space or '|' separated list of indexes that maps input frames to output
12379	frames. Number of indexes also sets maximal value that each index may have.
12380	'-1' index have special meaning and that is to drop frame.
12381	@end table
12382
12383	The first frame has the index 0. The default is to keep the input unchanged.
12384
12385	@subsection Examples
12386
12387	@itemize
12388	@item
12389	Swap second and third frame of every three frames of the input:
12390	@example
12391	ffmpeg -i INPUT -vf "shuffleframes=0 2 1" OUTPUT
12392	@end example
12393
12394	@item
12395	Swap 10th and 1st frame of every ten frames of the input:
12396	@example
12397	ffmpeg -i INPUT -vf "shuffleframes=9 1 2 3 4 5 6 7 8 0" OUTPUT
12398	@end example
12399	@end itemize
12400
12401	@section shuffleplanes
12402
12403	Reorder and/or duplicate video planes.
12404
12405	It accepts the following parameters:
12406
12407	@table @option
12408
12409	@item map0
12410	The index of the input plane to be used as the first output plane.
12411
12412	@item map1
12413	The index of the input plane to be used as the second output plane.
12414
12415	@item map2
12416	The index of the input plane to be used as the third output plane.
12417
12418	@item map3
12419	The index of the input plane to be used as the fourth output plane.
12420
12421	@end table
12422
12423	The first plane has the index 0. The default is to keep the input unchanged.
12424
12425	@subsection Examples
12426
12427	@itemize
12428	@item
12429	Swap the second and third planes of the input:
12430	@example
12431	ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
12432	@end example
12433	@end itemize
12434
12435	@anchor{signalstats}
12436	@section signalstats
12437	Evaluate various visual metrics that assist in determining issues associated
12438	with the digitization of analog video media.
12439
12440	By default the filter will log these metadata values:
12441
12442	@table @option
12443	@item YMIN
12444	Display the minimal Y value contained within the input frame. Expressed in
12445	range of [0-255].
12446
12447	@item YLOW
12448	Display the Y value at the 10% percentile within the input frame. Expressed in
12449	range of [0-255].
12450
12451	@item YAVG
12452	Display the average Y value within the input frame. Expressed in range of
12453	[0-255].
12454
12455	@item YHIGH
12456	Display the Y value at the 90% percentile within the input frame. Expressed in
12457	range of [0-255].
12458
12459	@item YMAX
12460	Display the maximum Y value contained within the input frame. Expressed in
12461	range of [0-255].
12462
12463	@item UMIN
12464	Display the minimal U value contained within the input frame. Expressed in
12465	range of [0-255].
12466
12467	@item ULOW
12468	Display the U value at the 10% percentile within the input frame. Expressed in
12469	range of [0-255].
12470
12471	@item UAVG
12472	Display the average U value within the input frame. Expressed in range of
12473	[0-255].
12474
12475	@item UHIGH
12476	Display the U value at the 90% percentile within the input frame. Expressed in
12477	range of [0-255].
12478
12479	@item UMAX
12480	Display the maximum U value contained within the input frame. Expressed in
12481	range of [0-255].
12482
12483	@item VMIN
12484	Display the minimal V value contained within the input frame. Expressed in
12485	range of [0-255].
12486
12487	@item VLOW
12488	Display the V value at the 10% percentile within the input frame. Expressed in
12489	range of [0-255].
12490
12491	@item VAVG
12492	Display the average V value within the input frame. Expressed in range of
12493	[0-255].
12494
12495	@item VHIGH
12496	Display the V value at the 90% percentile within the input frame. Expressed in
12497	range of [0-255].
12498
12499	@item VMAX
12500	Display the maximum V value contained within the input frame. Expressed in
12501	range of [0-255].
12502
12503	@item SATMIN
12504	Display the minimal saturation value contained within the input frame.
12505	Expressed in range of [0-~181.02].
12506
12507	@item SATLOW
12508	Display the saturation value at the 10% percentile within the input frame.
12509	Expressed in range of [0-~181.02].
12510
12511	@item SATAVG
12512	Display the average saturation value within the input frame. Expressed in range
12513	of [0-~181.02].
12514
12515	@item SATHIGH
12516	Display the saturation value at the 90% percentile within the input frame.
12517	Expressed in range of [0-~181.02].
12518
12519	@item SATMAX
12520	Display the maximum saturation value contained within the input frame.
12521	Expressed in range of [0-~181.02].
12522
12523	@item HUEMED
12524	Display the median value for hue within the input frame. Expressed in range of
12525	[0-360].
12526
12527	@item HUEAVG
12528	Display the average value for hue within the input frame. Expressed in range of
12529	[0-360].
12530
12531	@item YDIF
12532	Display the average of sample value difference between all values of the Y
12533	plane in the current frame and corresponding values of the previous input frame.
12534	Expressed in range of [0-255].
12535
12536	@item UDIF
12537	Display the average of sample value difference between all values of the U
12538	plane in the current frame and corresponding values of the previous input frame.
12539	Expressed in range of [0-255].
12540
12541	@item VDIF
12542	Display the average of sample value difference between all values of the V
12543	plane in the current frame and corresponding values of the previous input frame.
12544	Expressed in range of [0-255].
12545
12546	@item YBITDEPTH
12547	Display bit depth of Y plane in current frame.
12548	Expressed in range of [0-16].
12549
12550	@item UBITDEPTH
12551	Display bit depth of U plane in current frame.
12552	Expressed in range of [0-16].
12553
12554	@item VBITDEPTH
12555	Display bit depth of V plane in current frame.
12556	Expressed in range of [0-16].
12557	@end table
12558
12559	The filter accepts the following options:
12560
12561	@table @option
12562	@item stat
12563	@item out
12564
12565	@option{stat} specify an additional form of image analysis.
12566	@option{out} output video with the specified type of pixel highlighted.
12567
12568	Both options accept the following values:
12569
12570	@table @samp
12571	@item tout
12572	Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
12573	unlike the neighboring pixels of the same field. Examples of temporal outliers
12574	include the results of video dropouts, head clogs, or tape tracking issues.
12575
12576	@item vrep
12577	Identify @var{vertical line repetition}. Vertical line repetition includes
12578	similar rows of pixels within a frame. In born-digital video vertical line
12579	repetition is common, but this pattern is uncommon in video digitized from an
12580	analog source. When it occurs in video that results from the digitization of an
12581	analog source it can indicate concealment from a dropout compensator.
12582
12583	@item brng
12584	Identify pixels that fall outside of legal broadcast range.
12585	@end table
12586
12587	@item color, c
12588	Set the highlight color for the @option{out} option. The default color is
12589	yellow.
12590	@end table
12591
12592	@subsection Examples
12593
12594	@itemize
12595	@item
12596	Output data of various video metrics:
12597	@example
12598	ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
12599	@end example
12600
12601	@item
12602	Output specific data about the minimum and maximum values of the Y plane per frame:
12603	@example
12604	ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
12605	@end example
12606
12607	@item
12608	Playback video while highlighting pixels that are outside of broadcast range in red.
12609	@example
12610	ffplay example.mov -vf signalstats="out=brng:color=red"
12611	@end example
12612
12613	@item
12614	Playback video with signalstats metadata drawn over the frame.
12615	@example
12616	ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
12617	@end example
12618
12619	The contents of signalstat_drawtext.txt used in the command are:
12620	@example
12621	time %@{pts:hms@}
12622	Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
12623	U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
12624	V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
12625	saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
12626
12627	@end example
12628	@end itemize
12629
12630	@anchor{signature}
12631	@section signature
12632
12633	Calculates the MPEG-7 Video Signature. The filter can handle more than one
12634	input. In this case the matching between the inputs can be calculated additionally.
12635	The filter always passes through the first input. The signature of each stream can
12636	be written into a file.
12637
12638	It accepts the following options:
12639
12640	@table @option
12641	@item detectmode
12642	Enable or disable the matching process.
12643
12644	Available values are:
12645
12646	@table @samp
12647	@item off
12648	Disable the calculation of a matching (default).
12649	@item full
12650	Calculate the matching for the whole video and output whether the whole video
12651	matches or only parts.
12652	@item fast
12653	Calculate only until a matching is found or the video ends. Should be faster in
12654	some cases.
12655	@end table
12656
12657	@item nb_inputs
12658	Set the number of inputs. The option value must be a non negative integer.
12659	Default value is 1.
12660
12661	@item filename
12662	Set the path to which the output is written. If there is more than one input,
12663	the path must be a prototype, i.e. must contain %d or %0nd (where n is a positive
12664	integer), that will be replaced with the input number. If no filename is
12665	specified, no output will be written. This is the default.
12666
12667	@item format
12668	Choose the output format.
12669
12670	Available values are:
12671
12672	@table @samp
12673	@item binary
12674	Use the specified binary representation (default).
12675	@item xml
12676	Use the specified xml representation.
12677	@end table
12678
12679	@item th_d
12680	Set threshold to detect one word as similar. The option value must be an integer
12681	greater than zero. The default value is 9000.
12682
12683	@item th_dc
12684	Set threshold to detect all words as similar. The option value must be an integer
12685	greater than zero. The default value is 60000.
12686
12687	@item th_xh
12688	Set threshold to detect frames as similar. The option value must be an integer
12689	greater than zero. The default value is 116.
12690
12691	@item th_di
12692	Set the minimum length of a sequence in frames to recognize it as matching
12693	sequence. The option value must be a non negative integer value.
12694	The default value is 0.
12695
12696	@item th_it
12697	Set the minimum relation, that matching frames to all frames must have.
12698	The option value must be a double value between 0 and 1. The default value is 0.5.
12699	@end table
12700
12701	@subsection Examples
12702
12703	@itemize
12704	@item
12705	To calculate the signature of an input video and store it in signature.bin:
12706	@example
12707	ffmpeg -i input.mkv -vf signature=filename=signature.bin -map 0:v -f null -
12708	@end example
12709
12710	@item
12711	To detect whether two videos match and store the signatures in XML format in
12712	signature0.xml and signature1.xml:
12713	@example
12714	ffmpeg -i input1.mkv -i input2.mkv -filter_complex "[0:v][1:v] signature=nb_inputs=2:detectmode=full:format=xml:filename=signature%d.xml" -map :v -f null -
12715	@end example
12716
12717	@end itemize
12718
12719	@anchor{smartblur}
12720	@section smartblur
12721
12722	Blur the input video without impacting the outlines.
12723
12724	It accepts the following options:
12725
12726	@table @option
12727	@item luma_radius, lr
12728	Set the luma radius. The option value must be a float number in
12729	the range [0.1,5.0] that specifies the variance of the gaussian filter
12730	used to blur the image (slower if larger). Default value is 1.0.
12731
12732	@item luma_strength, ls
12733	Set the luma strength. The option value must be a float number
12734	in the range [-1.0,1.0] that configures the blurring. A value included
12735	in [0.0,1.0] will blur the image whereas a value included in
12736	[-1.0,0.0] will sharpen the image. Default value is 1.0.
12737
12738	@item luma_threshold, lt
12739	Set the luma threshold used as a coefficient to determine
12740	whether a pixel should be blurred or not. The option value must be an
12741	integer in the range [-30,30]. A value of 0 will filter all the image,
12742	a value included in [0,30] will filter flat areas and a value included
12743	in [-30,0] will filter edges. Default value is 0.
12744
12745	@item chroma_radius, cr
12746	Set the chroma radius. The option value must be a float number in
12747	the range [0.1,5.0] that specifies the variance of the gaussian filter
12748	used to blur the image (slower if larger). Default value is @option{luma_radius}.
12749
12750	@item chroma_strength, cs
12751	Set the chroma strength. The option value must be a float number
12752	in the range [-1.0,1.0] that configures the blurring. A value included
12753	in [0.0,1.0] will blur the image whereas a value included in
12754	[-1.0,0.0] will sharpen the image. Default value is @option{luma_strength}.
12755
12756	@item chroma_threshold, ct
12757	Set the chroma threshold used as a coefficient to determine
12758	whether a pixel should be blurred or not. The option value must be an
12759	integer in the range [-30,30]. A value of 0 will filter all the image,
12760	a value included in [0,30] will filter flat areas and a value included
12761	in [-30,0] will filter edges. Default value is @option{luma_threshold}.
12762	@end table
12763
12764	If a chroma option is not explicitly set, the corresponding luma value
12765	is set.
12766
12767	@section ssim
12768
12769	Obtain the SSIM (Structural SImilarity Metric) between two input videos.
12770
12771	This filter takes in input two input videos, the first input is
12772	considered the "main" source and is passed unchanged to the
12773	output. The second input is used as a "reference" video for computing
12774	the SSIM.
12775
12776	Both video inputs must have the same resolution and pixel format for
12777	this filter to work correctly. Also it assumes that both inputs
12778	have the same number of frames, which are compared one by one.
12779
12780	The filter stores the calculated SSIM of each frame.
12781
12782	The description of the accepted parameters follows.
12783
12784	@table @option
12785	@item stats_file, f
12786	If specified the filter will use the named file to save the SSIM of
12787	each individual frame. When filename equals "-" the data is sent to
12788	standard output.
12789	@end table
12790
12791	The file printed if @var{stats_file} is selected, contains a sequence of
12792	key/value pairs of the form @var{key}:@var{value} for each compared
12793	couple of frames.
12794
12795	A description of each shown parameter follows:
12796
12797	@table @option
12798	@item n
12799	sequential number of the input frame, starting from 1
12800
12801	@item Y, U, V, R, G, B
12802	SSIM of the compared frames for the component specified by the suffix.
12803
12804	@item All
12805	SSIM of the compared frames for the whole frame.
12806
12807	@item dB
12808	Same as above but in dB representation.
12809	@end table
12810
12811	For example:
12812	@example
12813	movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
12814	[main][ref] ssim="stats_file=stats.log" [out]
12815	@end example
12816
12817	On this example the input file being processed is compared with the
12818	reference file @file{ref_movie.mpg}. The SSIM of each individual frame
12819	is stored in @file{stats.log}.
12820
12821	Another example with both psnr and ssim at same time:
12822	@example
12823	ffmpeg -i main.mpg -i ref.mpg -lavfi "ssim;[0:v][1:v]psnr" -f null -
12824	@end example
12825
12826	@section stereo3d
12827
12828	Convert between different stereoscopic image formats.
12829
12830	The filters accept the following options:
12831
12832	@table @option
12833	@item in
12834	Set stereoscopic image format of input.
12835
12836	Available values for input image formats are:
12837	@table @samp
12838	@item sbsl
12839	side by side parallel (left eye left, right eye right)
12840
12841	@item sbsr
12842	side by side crosseye (right eye left, left eye right)
12843
12844	@item sbs2l
12845	side by side parallel with half width resolution
12846	(left eye left, right eye right)
12847
12848	@item sbs2r
12849	side by side crosseye with half width resolution
12850	(right eye left, left eye right)
12851
12852	@item abl
12853	above-below (left eye above, right eye below)
12854
12855	@item abr
12856	above-below (right eye above, left eye below)
12857
12858	@item ab2l
12859	above-below with half height resolution
12860	(left eye above, right eye below)
12861
12862	@item ab2r
12863	above-below with half height resolution
12864	(right eye above, left eye below)
12865
12866	@item al
12867	alternating frames (left eye first, right eye second)
12868
12869	@item ar
12870	alternating frames (right eye first, left eye second)
12871
12872	@item irl
12873	interleaved rows (left eye has top row, right eye starts on next row)
12874
12875	@item irr
12876	interleaved rows (right eye has top row, left eye starts on next row)
12877
12878	@item icl
12879	interleaved columns, left eye first
12880
12881	@item icr
12882	interleaved columns, right eye first
12883
12884	Default value is @samp{sbsl}.
12885	@end table
12886
12887	@item out
12888	Set stereoscopic image format of output.
12889
12890	@table @samp
12891	@item sbsl
12892	side by side parallel (left eye left, right eye right)
12893
12894	@item sbsr
12895	side by side crosseye (right eye left, left eye right)
12896
12897	@item sbs2l
12898	side by side parallel with half width resolution
12899	(left eye left, right eye right)
12900
12901	@item sbs2r
12902	side by side crosseye with half width resolution
12903	(right eye left, left eye right)
12904
12905	@item abl
12906	above-below (left eye above, right eye below)
12907
12908	@item abr
12909	above-below (right eye above, left eye below)
12910
12911	@item ab2l
12912	above-below with half height resolution
12913	(left eye above, right eye below)
12914
12915	@item ab2r
12916	above-below with half height resolution
12917	(right eye above, left eye below)
12918
12919	@item al
12920	alternating frames (left eye first, right eye second)
12921
12922	@item ar
12923	alternating frames (right eye first, left eye second)
12924
12925	@item irl
12926	interleaved rows (left eye has top row, right eye starts on next row)
12927
12928	@item irr
12929	interleaved rows (right eye has top row, left eye starts on next row)
12930
12931	@item arbg
12932	anaglyph red/blue gray
12933	(red filter on left eye, blue filter on right eye)
12934
12935	@item argg
12936	anaglyph red/green gray
12937	(red filter on left eye, green filter on right eye)
12938
12939	@item arcg
12940	anaglyph red/cyan gray
12941	(red filter on left eye, cyan filter on right eye)
12942
12943	@item arch
12944	anaglyph red/cyan half colored
12945	(red filter on left eye, cyan filter on right eye)
12946
12947	@item arcc
12948	anaglyph red/cyan color
12949	(red filter on left eye, cyan filter on right eye)
12950
12951	@item arcd
12952	anaglyph red/cyan color optimized with the least squares projection of dubois
12953	(red filter on left eye, cyan filter on right eye)
12954
12955	@item agmg
12956	anaglyph green/magenta gray
12957	(green filter on left eye, magenta filter on right eye)
12958
12959	@item agmh
12960	anaglyph green/magenta half colored
12961	(green filter on left eye, magenta filter on right eye)
12962
12963	@item agmc
12964	anaglyph green/magenta colored
12965	(green filter on left eye, magenta filter on right eye)
12966
12967	@item agmd
12968	anaglyph green/magenta color optimized with the least squares projection of dubois
12969	(green filter on left eye, magenta filter on right eye)
12970
12971	@item aybg
12972	anaglyph yellow/blue gray
12973	(yellow filter on left eye, blue filter on right eye)
12974
12975	@item aybh
12976	anaglyph yellow/blue half colored
12977	(yellow filter on left eye, blue filter on right eye)
12978
12979	@item aybc
12980	anaglyph yellow/blue colored
12981	(yellow filter on left eye, blue filter on right eye)
12982
12983	@item aybd
12984	anaglyph yellow/blue color optimized with the least squares projection of dubois
12985	(yellow filter on left eye, blue filter on right eye)
12986
12987	@item ml
12988	mono output (left eye only)
12989
12990	@item mr
12991	mono output (right eye only)
12992
12993	@item chl
12994	checkerboard, left eye first
12995
12996	@item chr
12997	checkerboard, right eye first
12998
12999	@item icl
13000	interleaved columns, left eye first
13001
13002	@item icr
13003	interleaved columns, right eye first
13004
13005	@item hdmi
13006	HDMI frame pack
13007	@end table
13008
13009	Default value is @samp{arcd}.
13010	@end table
13011
13012	@subsection Examples
13013
13014	@itemize
13015	@item
13016	Convert input video from side by side parallel to anaglyph yellow/blue dubois:
13017	@example
13018	stereo3d=sbsl:aybd
13019	@end example
13020
13021	@item
13022	Convert input video from above below (left eye above, right eye below) to side by side crosseye.
13023	@example
13024	stereo3d=abl:sbsr
13025	@end example
13026	@end itemize
13027
13028	@section streamselect, astreamselect
13029	Select video or audio streams.
13030
13031	The filter accepts the following options:
13032
13033	@table @option
13034	@item inputs
13035	Set number of inputs. Default is 2.
13036
13037	@item map
13038	Set input indexes to remap to outputs.
13039	@end table
13040
13041	@subsection Commands
13042
13043	The @code{streamselect} and @code{astreamselect} filter supports the following
13044	commands:
13045
13046	@table @option
13047	@item map
13048	Set input indexes to remap to outputs.
13049	@end table
13050
13051	@subsection Examples
13052
13053	@itemize
13054	@item
13055	Select first 5 seconds 1st stream and rest of time 2nd stream:
13056	@example
13057	sendcmd='5.0 streamselect map 1',streamselect=inputs=2:map=0
13058	@end example
13059
13060	@item
13061	Same as above, but for audio:
13062	@example
13063	asendcmd='5.0 astreamselect map 1',astreamselect=inputs=2:map=0
13064	@end example
13065	@end itemize
13066
13067	@section sobel
13068	Apply sobel operator to input video stream.
13069
13070	The filter accepts the following option:
13071
13072	@table @option
13073	@item planes
13074	Set which planes will be processed, unprocessed planes will be copied.
13075	By default value 0xf, all planes will be processed.
13076
13077	@item scale
13078	Set value which will be multiplied with filtered result.
13079
13080	@item delta
13081	Set value which will be added to filtered result.
13082	@end table
13083
13084	@anchor{spp}
13085	@section spp
13086
13087	Apply a simple postprocessing filter that compresses and decompresses the image
13088	at several (or - in the case of @option{quality} level @code{6} - all) shifts
13089	and average the results.
13090
13091	The filter accepts the following options:
13092
13093	@table @option
13094	@item quality
13095	Set quality. This option defines the number of levels for averaging. It accepts
13096	an integer in the range 0-6. If set to @code{0}, the filter will have no
13097	effect. A value of @code{6} means the higher quality. For each increment of
13098	that value the speed drops by a factor of approximately 2. Default value is
13099	@code{3}.
13100
13101	@item qp
13102	Force a constant quantization parameter. If not set, the filter will use the QP
13103	from the video stream (if available).
13104
13105	@item mode
13106	Set thresholding mode. Available modes are:
13107
13108	@table @samp
13109	@item hard
13110	Set hard thresholding (default).
13111	@item soft
13112	Set soft thresholding (better de-ringing effect, but likely blurrier).
13113	@end table
13114
13115	@item use_bframe_qp
13116	Enable the use of the QP from the B-Frames if set to @code{1}. Using this
13117	option may cause flicker since the B-Frames have often larger QP. Default is
13118	@code{0} (not enabled).
13119	@end table
13120
13121	@anchor{subtitles}
13122	@section subtitles
13123
13124	Draw subtitles on top of input video using the libass library.
13125
13126	To enable compilation of this filter you need to configure FFmpeg with
13127	@code{--enable-libass}. This filter also requires a build with libavcodec and
13128	libavformat to convert the passed subtitles file to ASS (Advanced Substation
13129	Alpha) subtitles format.
13130
13131	The filter accepts the following options:
13132
13133	@table @option
13134	@item filename, f
13135	Set the filename of the subtitle file to read. It must be specified.
13136
13137	@item original_size
13138	Specify the size of the original video, the video for which the ASS file
13139	was composed. For the syntax of this option, check the
13140	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13141	Due to a misdesign in ASS aspect ratio arithmetic, this is necessary to
13142	correctly scale the fonts if the aspect ratio has been changed.
13143
13144	@item fontsdir
13145	Set a directory path containing fonts that can be used by the filter.
13146	These fonts will be used in addition to whatever the font provider uses.
13147
13148	@item charenc
13149	Set subtitles input character encoding. @code{subtitles} filter only. Only
13150	useful if not UTF-8.
13151
13152	@item stream_index, si
13153	Set subtitles stream index. @code{subtitles} filter only.
13154
13155	@item force_style
13156	Override default style or script info parameters of the subtitles. It accepts a
13157	string containing ASS style format @code{KEY=VALUE} couples separated by ",".
13158	@end table
13159
13160	If the first key is not specified, it is assumed that the first value
13161	specifies the @option{filename}.
13162
13163	For example, to render the file @file{sub.srt} on top of the input
13164	video, use the command:
13165	@example
13166	subtitles=sub.srt
13167	@end example
13168
13169	which is equivalent to:
13170	@example
13171	subtitles=filename=sub.srt
13172	@end example
13173
13174	To render the default subtitles stream from file @file{video.mkv}, use:
13175	@example
13176	subtitles=video.mkv
13177	@end example
13178
13179	To render the second subtitles stream from that file, use:
13180	@example
13181	subtitles=video.mkv:si=1
13182	@end example
13183
13184	To make the subtitles stream from @file{sub.srt} appear in transparent green
13185	@code{DejaVu Serif}, use:
13186	@example
13187	subtitles=sub.srt:force_style='FontName=DejaVu Serif,PrimaryColour=&HAA00FF00'
13188	@end example
13189
13190	@section super2xsai
13191
13192	Scale the input by 2x and smooth using the Super2xSaI (Scale and
13193	Interpolate) pixel art scaling algorithm.
13194
13195	Useful for enlarging pixel art images without reducing sharpness.
13196
13197	@section swaprect
13198
13199	Swap two rectangular objects in video.
13200
13201	This filter accepts the following options:
13202
13203	@table @option
13204	@item w
13205	Set object width.
13206
13207	@item h
13208	Set object height.
13209
13210	@item x1
13211	Set 1st rect x coordinate.
13212
13213	@item y1
13214	Set 1st rect y coordinate.
13215
13216	@item x2
13217	Set 2nd rect x coordinate.
13218
13219	@item y2
13220	Set 2nd rect y coordinate.
13221
13222	All expressions are evaluated once for each frame.
13223	@end table
13224
13225	The all options are expressions containing the following constants:
13226
13227	@table @option
13228	@item w
13229	@item h
13230	The input width and height.
13231
13232	@item a
13233	same as @var{w} / @var{h}
13234
13235	@item sar
13236	input sample aspect ratio
13237
13238	@item dar
13239	input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
13240
13241	@item n
13242	The number of the input frame, starting from 0.
13243
13244	@item t
13245	The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
13246
13247	@item pos
13248	the position in the file of the input frame, NAN if unknown
13249	@end table
13250
13251	@section swapuv
13252	Swap U & V plane.
13253
13254	@section telecine
13255
13256	Apply telecine process to the video.
13257
13258	This filter accepts the following options:
13259
13260	@table @option
13261	@item first_field
13262	@table @samp
13263	@item top, t
13264	top field first
13265	@item bottom, b
13266	bottom field first
13267	The default value is @code{top}.
13268	@end table
13269
13270	@item pattern
13271	A string of numbers representing the pulldown pattern you wish to apply.
13272	The default value is @code{23}.
13273	@end table
13274
13275	@example
13276	Some typical patterns:
13277
13278	NTSC output (30i):
13279	27.5p: 32222
13280	24p: 23 (classic)
13281	24p: 2332 (preferred)
13282	20p: 33
13283	18p: 334
13284	16p: 3444
13285
13286	PAL output (25i):
13287	27.5p: 12222
13288	24p: 222222222223 ("Euro pulldown")
13289	16.67p: 33
13290	16p: 33333334
13291	@end example
13292
13293	@section threshold
13294
13295	Apply threshold effect to video stream.
13296
13297	This filter needs four video streams to perform thresholding.
13298	First stream is stream we are filtering.
13299	Second stream is holding threshold values, third stream is holding min values,
13300	and last, fourth stream is holding max values.
13301
13302	The filter accepts the following option:
13303
13304	@table @option
13305	@item planes
13306	Set which planes will be processed, unprocessed planes will be copied.
13307	By default value 0xf, all planes will be processed.
13308	@end table
13309
13310	For example if first stream pixel's component value is less then threshold value
13311	of pixel component from 2nd threshold stream, third stream value will picked,
13312	otherwise fourth stream pixel component value will be picked.
13313
13314	Using color source filter one can perform various types of thresholding:
13315
13316	@subsection Examples
13317
13318	@itemize
13319	@item
13320	Binary threshold, using gray color as threshold:
13321	@example
13322	ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=black -f lavfi -i color=white -lavfi threshold output.avi
13323	@end example
13324
13325	@item
13326	Inverted binary threshold, using gray color as threshold:
13327	@example
13328	ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -f lavfi -i color=black -lavfi threshold output.avi
13329	@end example
13330
13331	@item
13332	Truncate binary threshold, using gray color as threshold:
13333	@example
13334	ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=gray -lavfi threshold output.avi
13335	@end example
13336
13337	@item
13338	Threshold to zero, using gray color as threshold:
13339	@example
13340	ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -i 320x240.avi -lavfi threshold output.avi
13341	@end example
13342
13343	@item
13344	Inverted threshold to zero, using gray color as threshold:
13345	@example
13346	ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=white -lavfi threshold output.avi
13347	@end example
13348	@end itemize
13349
13350	@section thumbnail
13351	Select the most representative frame in a given sequence of consecutive frames.
13352
13353	The filter accepts the following options:
13354
13355	@table @option
13356	@item n
13357	Set the frames batch size to analyze; in a set of @var{n} frames, the filter
13358	will pick one of them, and then handle the next batch of @var{n} frames until
13359	the end. Default is @code{100}.
13360	@end table
13361
13362	Since the filter keeps track of the whole frames sequence, a bigger @var{n}
13363	value will result in a higher memory usage, so a high value is not recommended.
13364
13365	@subsection Examples
13366
13367	@itemize
13368	@item
13369	Extract one picture each 50 frames:
13370	@example
13371	thumbnail=50
13372	@end example
13373
13374	@item
13375	Complete example of a thumbnail creation with @command{ffmpeg}:
13376	@example
13377	ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
13378	@end example
13379	@end itemize
13380
13381	@section tile
13382
13383	Tile several successive frames together.
13384
13385	The filter accepts the following options:
13386
13387	@table @option
13388
13389	@item layout
13390	Set the grid size (i.e. the number of lines and columns). For the syntax of
13391	this option, check the
13392	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13393
13394	@item nb_frames
13395	Set the maximum number of frames to render in the given area. It must be less
13396	than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
13397	the area will be used.
13398
13399	@item margin
13400	Set the outer border margin in pixels.
13401
13402	@item padding
13403	Set the inner border thickness (i.e. the number of pixels between frames). For
13404	more advanced padding options (such as having different values for the edges),
13405	refer to the pad video filter.
13406
13407	@item color
13408	Specify the color of the unused area. For the syntax of this option, check the
13409	"Color" section in the ffmpeg-utils manual. The default value of @var{color}
13410	is "black".
13411	@end table
13412
13413	@subsection Examples
13414
13415	@itemize
13416	@item
13417	Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
13418	@example
13419	ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
13420	@end example
13421	The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
13422	duplicating each output frame to accommodate the originally detected frame
13423	rate.
13424
13425	@item
13426	Display @code{5} pictures in an area of @code{3x2} frames,
13427	with @code{7} pixels between them, and @code{2} pixels of initial margin, using
13428	mixed flat and named options:
13429	@example
13430	tile=3x2:nb_frames=5:padding=7:margin=2
13431	@end example
13432	@end itemize
13433
13434	@section tinterlace
13435
13436	Perform various types of temporal field interlacing.
13437
13438	Frames are counted starting from 1, so the first input frame is
13439	considered odd.
13440
13441	The filter accepts the following options:
13442
13443	@table @option
13444
13445	@item mode
13446	Specify the mode of the interlacing. This option can also be specified
13447	as a value alone. See below for a list of values for this option.
13448
13449	Available values are:
13450
13451	@table @samp
13452	@item merge, 0
13453	Move odd frames into the upper field, even into the lower field,
13454	generating a double height frame at half frame rate.
13455	@example
13456	------> time
13457	Input:
13458	Frame 1 Frame 2 Frame 3 Frame 4
13459
13460	11111 22222 33333 44444
13461	11111 22222 33333 44444
13462	11111 22222 33333 44444
13463	11111 22222 33333 44444
13464
13465	Output:
13466	11111 33333
13467	22222 44444
13468	11111 33333
13469	22222 44444
13470	11111 33333
13471	22222 44444
13472	11111 33333
13473	22222 44444
13474	@end example
13475
13476	@item drop_even, 1
13477	Only output odd frames, even frames are dropped, generating a frame with
13478	unchanged height at half frame rate.
13479
13480	@example
13481	------> time
13482	Input:
13483	Frame 1 Frame 2 Frame 3 Frame 4
13484
13485	11111 22222 33333 44444
13486	11111 22222 33333 44444
13487	11111 22222 33333 44444
13488	11111 22222 33333 44444
13489
13490	Output:
13491	11111 33333
13492	11111 33333
13493	11111 33333
13494	11111 33333
13495	@end example
13496
13497	@item drop_odd, 2
13498	Only output even frames, odd frames are dropped, generating a frame with
13499	unchanged height at half frame rate.
13500
13501	@example
13502	------> time
13503	Input:
13504	Frame 1 Frame 2 Frame 3 Frame 4
13505
13506	11111 22222 33333 44444
13507	11111 22222 33333 44444
13508	11111 22222 33333 44444
13509	11111 22222 33333 44444
13510
13511	Output:
13512	22222 44444
13513	22222 44444
13514	22222 44444
13515	22222 44444
13516	@end example
13517
13518	@item pad, 3
13519	Expand each frame to full height, but pad alternate lines with black,
13520	generating a frame with double height at the same input frame rate.
13521
13522	@example
13523	------> time
13524	Input:
13525	Frame 1 Frame 2 Frame 3 Frame 4
13526
13527	11111 22222 33333 44444
13528	11111 22222 33333 44444
13529	11111 22222 33333 44444
13530	11111 22222 33333 44444
13531
13532	Output:
13533	11111 ..... 33333 .....
13534	..... 22222 ..... 44444
13535	11111 ..... 33333 .....
13536	..... 22222 ..... 44444
13537	11111 ..... 33333 .....
13538	..... 22222 ..... 44444
13539	11111 ..... 33333 .....
13540	..... 22222 ..... 44444
13541	@end example
13542
13543
13544	@item interleave_top, 4
13545	Interleave the upper field from odd frames with the lower field from
13546	even frames, generating a frame with unchanged height at half frame rate.
13547
13548	@example
13549	------> time
13550	Input:
13551	Frame 1 Frame 2 Frame 3 Frame 4
13552
13553	11111<- 22222 33333<- 44444
13554	11111 22222<- 33333 44444<-
13555	11111<- 22222 33333<- 44444
13556	11111 22222<- 33333 44444<-
13557
13558	Output:
13559	11111 33333
13560	22222 44444
13561	11111 33333
13562	22222 44444
13563	@end example
13564
13565
13566	@item interleave_bottom, 5
13567	Interleave the lower field from odd frames with the upper field from
13568	even frames, generating a frame with unchanged height at half frame rate.
13569
13570	@example
13571	------> time
13572	Input:
13573	Frame 1 Frame 2 Frame 3 Frame 4
13574
13575	11111 22222<- 33333 44444<-
13576	11111<- 22222 33333<- 44444
13577	11111 22222<- 33333 44444<-
13578	11111<- 22222 33333<- 44444
13579
13580	Output:
13581	22222 44444
13582	11111 33333
13583	22222 44444
13584	11111 33333
13585	@end example
13586
13587
13588	@item interlacex2, 6
13589	Double frame rate with unchanged height. Frames are inserted each
13590	containing the second temporal field from the previous input frame and
13591	the first temporal field from the next input frame. This mode relies on
13592	the top_field_first flag. Useful for interlaced video displays with no
13593	field synchronisation.
13594
13595	@example
13596	------> time
13597	Input:
13598	Frame 1 Frame 2 Frame 3 Frame 4
13599
13600	11111 22222 33333 44444
13601	11111 22222 33333 44444
13602	11111 22222 33333 44444
13603	11111 22222 33333 44444
13604
13605	Output:
13606	11111 22222 22222 33333 33333 44444 44444
13607	11111 11111 22222 22222 33333 33333 44444
13608	11111 22222 22222 33333 33333 44444 44444
13609	11111 11111 22222 22222 33333 33333 44444
13610	@end example
13611
13612
13613	@item mergex2, 7
13614	Move odd frames into the upper field, even into the lower field,
13615	generating a double height frame at same frame rate.
13616
13617	@example
13618	------> time
13619	Input:
13620	Frame 1 Frame 2 Frame 3 Frame 4
13621
13622	11111 22222 33333 44444
13623	11111 22222 33333 44444
13624	11111 22222 33333 44444
13625	11111 22222 33333 44444
13626
13627	Output:
13628	11111 33333 33333 55555
13629	22222 22222 44444 44444
13630	11111 33333 33333 55555
13631	22222 22222 44444 44444
13632	11111 33333 33333 55555
13633	22222 22222 44444 44444
13634	11111 33333 33333 55555
13635	22222 22222 44444 44444
13636	@end example
13637
13638	@end table
13639
13640	Numeric values are deprecated but are accepted for backward
13641	compatibility reasons.
13642
13643	Default mode is @code{merge}.
13644
13645	@item flags
13646	Specify flags influencing the filter process.
13647
13648	Available value for @var{flags} is:
13649
13650	@table @option
13651	@item low_pass_filter, vlfp
13652	Enable vertical low-pass filtering in the filter.
13653	Vertical low-pass filtering is required when creating an interlaced
13654	destination from a progressive source which contains high-frequency
13655	vertical detail. Filtering will reduce interlace 'twitter' and Moire
13656	patterning.
13657
13658	Vertical low-pass filtering can only be enabled for @option{mode}
13659	@var{interleave_top} and @var{interleave_bottom}.
13660
13661	@end table
13662	@end table
13663
13664	@section transpose
13665
13666	Transpose rows with columns in the input video and optionally flip it.
13667
13668	It accepts the following parameters:
13669
13670	@table @option
13671
13672	@item dir
13673	Specify the transposition direction.
13674
13675	Can assume the following values:
13676	@table @samp
13677	@item 0, 4, cclock_flip
13678	Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
13679	@example
13680	L.R L.l
13681	. . -> . .
13682	l.r R.r
13683	@end example
13684
13685	@item 1, 5, clock
13686	Rotate by 90 degrees clockwise, that is:
13687	@example
13688	L.R l.L
13689	. . -> . .
13690	l.r r.R
13691	@end example
13692
13693	@item 2, 6, cclock
13694	Rotate by 90 degrees counterclockwise, that is:
13695	@example
13696	L.R R.r
13697	. . -> . .
13698	l.r L.l
13699	@end example
13700
13701	@item 3, 7, clock_flip
13702	Rotate by 90 degrees clockwise and vertically flip, that is:
13703	@example
13704	L.R r.R
13705	. . -> . .
13706	l.r l.L
13707	@end example
13708	@end table
13709
13710	For values between 4-7, the transposition is only done if the input
13711	video geometry is portrait and not landscape. These values are
13712	deprecated, the @code{passthrough} option should be used instead.
13713
13714	Numerical values are deprecated, and should be dropped in favor of
13715	symbolic constants.
13716
13717	@item passthrough
13718	Do not apply the transposition if the input geometry matches the one
13719	specified by the specified value. It accepts the following values:
13720	@table @samp
13721	@item none
13722	Always apply transposition.
13723	@item portrait
13724	Preserve portrait geometry (when @var{height} >= @var{width}).
13725	@item landscape
13726	Preserve landscape geometry (when @var{width} >= @var{height}).
13727	@end table
13728
13729	Default value is @code{none}.
13730	@end table
13731
13732	For example to rotate by 90 degrees clockwise and preserve portrait
13733	layout:
13734	@example
13735	transpose=dir=1:passthrough=portrait
13736	@end example
13737
13738	The command above can also be specified as:
13739	@example
13740	transpose=1:portrait
13741	@end example
13742
13743	@section trim
13744	Trim the input so that the output contains one continuous subpart of the input.
13745
13746	It accepts the following parameters:
13747	@table @option
13748	@item start
13749	Specify the time of the start of the kept section, i.e. the frame with the
13750	timestamp @var{start} will be the first frame in the output.
13751
13752	@item end
13753	Specify the time of the first frame that will be dropped, i.e. the frame
13754	immediately preceding the one with the timestamp @var{end} will be the last
13755	frame in the output.
13756
13757	@item start_pts
13758	This is the same as @var{start}, except this option sets the start timestamp
13759	in timebase units instead of seconds.
13760
13761	@item end_pts
13762	This is the same as @var{end}, except this option sets the end timestamp
13763	in timebase units instead of seconds.
13764
13765	@item duration
13766	The maximum duration of the output in seconds.
13767
13768	@item start_frame
13769	The number of the first frame that should be passed to the output.
13770
13771	@item end_frame
13772	The number of the first frame that should be dropped.
13773	@end table
13774
13775	@option{start}, @option{end}, and @option{duration} are expressed as time
13776	duration specifications; see
13777	@ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
13778	for the accepted syntax.
13779
13780	Note that the first two sets of the start/end options and the @option{duration}
13781	option look at the frame timestamp, while the _frame variants simply count the
13782	frames that pass through the filter. Also note that this filter does not modify
13783	the timestamps. If you wish for the output timestamps to start at zero, insert a
13784	setpts filter after the trim filter.
13785
13786	If multiple start or end options are set, this filter tries to be greedy and
13787	keep all the frames that match at least one of the specified constraints. To keep
13788	only the part that matches all the constraints at once, chain multiple trim
13789	filters.
13790
13791	The defaults are such that all the input is kept. So it is possible to set e.g.
13792	just the end values to keep everything before the specified time.
13793
13794	Examples:
13795	@itemize
13796	@item
13797	Drop everything except the second minute of input:
13798	@example
13799	ffmpeg -i INPUT -vf trim=60:120
13800	@end example
13801
13802	@item
13803	Keep only the first second:
13804	@example
13805	ffmpeg -i INPUT -vf trim=duration=1
13806	@end example
13807
13808	@end itemize
13809
13810
13811	@anchor{unsharp}
13812	@section unsharp
13813
13814	Sharpen or blur the input video.
13815
13816	It accepts the following parameters:
13817
13818	@table @option
13819	@item luma_msize_x, lx
13820	Set the luma matrix horizontal size. It must be an odd integer between
13821	3 and 23. The default value is 5.
13822
13823	@item luma_msize_y, ly
13824	Set the luma matrix vertical size. It must be an odd integer between 3
13825	and 23. The default value is 5.
13826
13827	@item luma_amount, la
13828	Set the luma effect strength. It must be a floating point number, reasonable
13829	values lay between -1.5 and 1.5.
13830
13831	Negative values will blur the input video, while positive values will
13832	sharpen it, a value of zero will disable the effect.
13833
13834	Default value is 1.0.
13835
13836	@item chroma_msize_x, cx
13837	Set the chroma matrix horizontal size. It must be an odd integer
13838	between 3 and 23. The default value is 5.
13839
13840	@item chroma_msize_y, cy
13841	Set the chroma matrix vertical size. It must be an odd integer
13842	between 3 and 23. The default value is 5.
13843
13844	@item chroma_amount, ca
13845	Set the chroma effect strength. It must be a floating point number, reasonable
13846	values lay between -1.5 and 1.5.
13847
13848	Negative values will blur the input video, while positive values will
13849	sharpen it, a value of zero will disable the effect.
13850
13851	Default value is 0.0.
13852
13853	@item opencl
13854	If set to 1, specify using OpenCL capabilities, only available if
13855	FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
13856
13857	@end table
13858
13859	All parameters are optional and default to the equivalent of the
13860	string '5:5:1.0:5:5:0.0'.
13861
13862	@subsection Examples
13863
13864	@itemize
13865	@item
13866	Apply strong luma sharpen effect:
13867	@example
13868	unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
13869	@end example
13870
13871	@item
13872	Apply a strong blur of both luma and chroma parameters:
13873	@example
13874	unsharp=7:7:-2:7:7:-2
13875	@end example
13876	@end itemize
13877
13878	@section uspp
13879
13880	Apply ultra slow/simple postprocessing filter that compresses and decompresses
13881	the image at several (or - in the case of @option{quality} level @code{8} - all)
13882	shifts and average the results.
13883
13884	The way this differs from the behavior of spp is that uspp actually encodes &
13885	decodes each case with libavcodec Snow, whereas spp uses a simplified intra only 8x8
13886	DCT similar to MJPEG.
13887
13888	The filter accepts the following options:
13889
13890	@table @option
13891	@item quality
13892	Set quality. This option defines the number of levels for averaging. It accepts
13893	an integer in the range 0-8. If set to @code{0}, the filter will have no
13894	effect. A value of @code{8} means the higher quality. For each increment of
13895	that value the speed drops by a factor of approximately 2. Default value is
13896	@code{3}.
13897
13898	@item qp
13899	Force a constant quantization parameter. If not set, the filter will use the QP
13900	from the video stream (if available).
13901	@end table
13902
13903	@section vaguedenoiser
13904
13905	Apply a wavelet based denoiser.
13906
13907	It transforms each frame from the video input into the wavelet domain,
13908	using Cohen-Daubechies-Feauveau 9/7. Then it applies some filtering to
13909	the obtained coefficients. It does an inverse wavelet transform after.
13910	Due to wavelet properties, it should give a nice smoothed result, and
13911	reduced noise, without blurring picture features.
13912
13913	This filter accepts the following options:
13914
13915	@table @option
13916	@item threshold
13917	The filtering strength. The higher, the more filtered the video will be.
13918	Hard thresholding can use a higher threshold than soft thresholding
13919	before the video looks overfiltered.
13920
13921	@item method
13922	The filtering method the filter will use.
13923
13924	It accepts the following values:
13925	@table @samp
13926	@item hard
13927	All values under the threshold will be zeroed.
13928
13929	@item soft
13930	All values under the threshold will be zeroed. All values above will be
13931	reduced by the threshold.
13932
13933	@item garrote
13934	Scales or nullifies coefficients - intermediary between (more) soft and
13935	(less) hard thresholding.
13936	@end table
13937
13938	@item nsteps
13939	Number of times, the wavelet will decompose the picture. Picture can't
13940	be decomposed beyond a particular point (typically, 8 for a 640x480
13941	frame - as 2^9 = 512 > 480)
13942
13943	@item percent
13944	Partial of full denoising (limited coefficients shrinking), from 0 to 100.
13945
13946	@item planes
13947	A list of the planes to process. By default all planes are processed.
13948	@end table
13949
13950	@section vectorscope
13951
13952	Display 2 color component values in the two dimensional graph (which is called
13953	a vectorscope).
13954
13955	This filter accepts the following options:
13956
13957	@table @option
13958	@item mode, m
13959	Set vectorscope mode.
13960
13961	It accepts the following values:
13962	@table @samp
13963	@item gray
13964	Gray values are displayed on graph, higher brightness means more pixels have
13965	same component color value on location in graph. This is the default mode.
13966
13967	@item color
13968	Gray values are displayed on graph. Surrounding pixels values which are not
13969	present in video frame are drawn in gradient of 2 color components which are
13970	set by option @code{x} and @code{y}. The 3rd color component is static.
13971
13972	@item color2
13973	Actual color components values present in video frame are displayed on graph.
13974
13975	@item color3
13976	Similar as color2 but higher frequency of same values @code{x} and @code{y}
13977	on graph increases value of another color component, which is luminance by
13978	default values of @code{x} and @code{y}.
13979
13980	@item color4
13981	Actual colors present in video frame are displayed on graph. If two different
13982	colors map to same position on graph then color with higher value of component
13983	not present in graph is picked.
13984
13985	@item color5
13986	Gray values are displayed on graph. Similar to @code{color} but with 3rd color
13987	component picked from radial gradient.
13988	@end table
13989
13990	@item x
13991	Set which color component will be represented on X-axis. Default is @code{1}.
13992
13993	@item y
13994	Set which color component will be represented on Y-axis. Default is @code{2}.
13995
13996	@item intensity, i
13997	Set intensity, used by modes: gray, color, color3 and color5 for increasing brightness
13998	of color component which represents frequency of (X, Y) location in graph.
13999
14000	@item envelope, e
14001	@table @samp
14002	@item none
14003	No envelope, this is default.
14004
14005	@item instant
14006	Instant envelope, even darkest single pixel will be clearly highlighted.
14007
14008	@item peak
14009	Hold maximum and minimum values presented in graph over time. This way you
14010	can still spot out of range values without constantly looking at vectorscope.
14011
14012	@item peak+instant
14013	Peak and instant envelope combined together.
14014	@end table
14015
14016	@item graticule, g
14017	Set what kind of graticule to draw.
14018	@table @samp
14019	@item none
14020	@item green
14021	@item color
14022	@end table
14023
14024	@item opacity, o
14025	Set graticule opacity.
14026
14027	@item flags, f
14028	Set graticule flags.
14029
14030	@table @samp
14031	@item white
14032	Draw graticule for white point.
14033
14034	@item black
14035	Draw graticule for black point.
14036
14037	@item name
14038	Draw color points short names.
14039	@end table
14040
14041	@item bgopacity, b
14042	Set background opacity.
14043
14044	@item lthreshold, l
14045	Set low threshold for color component not represented on X or Y axis.
14046	Values lower than this value will be ignored. Default is 0.
14047	Note this value is multiplied with actual max possible value one pixel component
14048	can have. So for 8-bit input and low threshold value of 0.1 actual threshold
14049	is 0.1 * 255 = 25.
14050
14051	@item hthreshold, h
14052	Set high threshold for color component not represented on X or Y axis.
14053	Values higher than this value will be ignored. Default is 1.
14054	Note this value is multiplied with actual max possible value one pixel component
14055	can have. So for 8-bit input and high threshold value of 0.9 actual threshold
14056	is 0.9 * 255 = 230.
14057
14058	@item colorspace, c
14059	Set what kind of colorspace to use when drawing graticule.
14060	@table @samp
14061	@item auto
14062	@item 601
14063	@item 709
14064	@end table
14065	Default is auto.
14066	@end table
14067
14068	@anchor{vidstabdetect}
14069	@section vidstabdetect
14070
14071	Analyze video stabilization/deshaking. Perform pass 1 of 2, see
14072	@ref{vidstabtransform} for pass 2.
14073
14074	This filter generates a file with relative translation and rotation
14075	transform information about subsequent frames, which is then used by
14076	the @ref{vidstabtransform} filter.
14077
14078	To enable compilation of this filter you need to configure FFmpeg with
14079	@code{--enable-libvidstab}.
14080
14081	This filter accepts the following options:
14082
14083	@table @option
14084	@item result
14085	Set the path to the file used to write the transforms information.
14086	Default value is @file{transforms.trf}.
14087
14088	@item shakiness
14089	Set how shaky the video is and how quick the camera is. It accepts an
14090	integer in the range 1-10, a value of 1 means little shakiness, a
14091	value of 10 means strong shakiness. Default value is 5.
14092
14093	@item accuracy
14094	Set the accuracy of the detection process. It must be a value in the
14095	range 1-15. A value of 1 means low accuracy, a value of 15 means high
14096	accuracy. Default value is 15.
14097
14098	@item stepsize
14099	Set stepsize of the search process. The region around minimum is
14100	scanned with 1 pixel resolution. Default value is 6.
14101
14102	@item mincontrast
14103	Set minimum contrast. Below this value a local measurement field is
14104	discarded. Must be a floating point value in the range 0-1. Default
14105	value is 0.3.
14106
14107	@item tripod
14108	Set reference frame number for tripod mode.
14109
14110	If enabled, the motion of the frames is compared to a reference frame
14111	in the filtered stream, identified by the specified number. The idea
14112	is to compensate all movements in a more-or-less static scene and keep
14113	the camera view absolutely still.
14114
14115	If set to 0, it is disabled. The frames are counted starting from 1.
14116
14117	@item show
14118	Show fields and transforms in the resulting frames. It accepts an
14119	integer in the range 0-2. Default value is 0, which disables any
14120	visualization.
14121	@end table
14122
14123	@subsection Examples
14124
14125	@itemize
14126	@item
14127	Use default values:
14128	@example
14129	vidstabdetect
14130	@end example
14131
14132	@item
14133	Analyze strongly shaky movie and put the results in file
14134	@file{mytransforms.trf}:
14135	@example
14136	vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
14137	@end example
14138
14139	@item
14140	Visualize the result of internal transformations in the resulting
14141	video:
14142	@example
14143	vidstabdetect=show=1
14144	@end example
14145
14146	@item
14147	Analyze a video with medium shakiness using @command{ffmpeg}:
14148	@example
14149	ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
14150	@end example
14151	@end itemize
14152
14153	@anchor{vidstabtransform}
14154	@section vidstabtransform
14155
14156	Video stabilization/deshaking: pass 2 of 2,
14157	see @ref{vidstabdetect} for pass 1.
14158
14159	Read a file with transform information for each frame and
14160	apply/compensate them. Together with the @ref{vidstabdetect}
14161	filter this can be used to deshake videos. See also
14162	@url{http://public.hronopik.de/vid.stab}. It is important to also use
14163	the @ref{unsharp} filter, see below.
14164
14165	To enable compilation of this filter you need to configure FFmpeg with
14166	@code{--enable-libvidstab}.
14167
14168	@subsection Options
14169
14170	@table @option
14171	@item input
14172	Set path to the file used to read the transforms. Default value is
14173	@file{transforms.trf}.
14174
14175	@item smoothing
14176	Set the number of frames (value*2 + 1) used for lowpass filtering the
14177	camera movements. Default value is 10.
14178
14179	For example a number of 10 means that 21 frames are used (10 in the
14180	past and 10 in the future) to smoothen the motion in the video. A
14181	larger value leads to a smoother video, but limits the acceleration of
14182	the camera (pan/tilt movements). 0 is a special case where a static
14183	camera is simulated.
14184
14185	@item optalgo
14186	Set the camera path optimization algorithm.
14187
14188	Accepted values are:
14189	@table @samp
14190	@item gauss
14191	gaussian kernel low-pass filter on camera motion (default)
14192	@item avg
14193	averaging on transformations
14194	@end table
14195
14196	@item maxshift
14197	Set maximal number of pixels to translate frames. Default value is -1,
14198	meaning no limit.
14199
14200	@item maxangle
14201	Set maximal angle in radians (degree*PI/180) to rotate frames. Default
14202	value is -1, meaning no limit.
14203
14204	@item crop
14205	Specify how to deal with borders that may be visible due to movement
14206	compensation.
14207
14208	Available values are:
14209	@table @samp
14210	@item keep
14211	keep image information from previous frame (default)
14212	@item black
14213	fill the border black
14214	@end table
14215
14216	@item invert
14217	Invert transforms if set to 1. Default value is 0.
14218
14219	@item relative
14220	Consider transforms as relative to previous frame if set to 1,
14221	absolute if set to 0. Default value is 0.
14222
14223	@item zoom
14224	Set percentage to zoom. A positive value will result in a zoom-in
14225	effect, a negative value in a zoom-out effect. Default value is 0 (no
14226	zoom).
14227
14228	@item optzoom
14229	Set optimal zooming to avoid borders.
14230
14231	Accepted values are:
14232	@table @samp
14233	@item 0
14234	disabled
14235	@item 1
14236	optimal static zoom value is determined (only very strong movements
14237	will lead to visible borders) (default)
14238	@item 2
14239	optimal adaptive zoom value is determined (no borders will be
14240	visible), see @option{zoomspeed}
14241	@end table
14242
14243	Note that the value given at zoom is added to the one calculated here.
14244
14245	@item zoomspeed
14246	Set percent to zoom maximally each frame (enabled when
14247	@option{optzoom} is set to 2). Range is from 0 to 5, default value is
14248	0.25.
14249
14250	@item interpol
14251	Specify type of interpolation.
14252
14253	Available values are:
14254	@table @samp
14255	@item no
14256	no interpolation
14257	@item linear
14258	linear only horizontal
14259	@item bilinear
14260	linear in both directions (default)
14261	@item bicubic
14262	cubic in both directions (slow)
14263	@end table
14264
14265	@item tripod
14266	Enable virtual tripod mode if set to 1, which is equivalent to
14267	@code{relative=0:smoothing=0}. Default value is 0.
14268
14269	Use also @code{tripod} option of @ref{vidstabdetect}.
14270
14271	@item debug
14272	Increase log verbosity if set to 1. Also the detected global motions
14273	are written to the temporary file @file{global_motions.trf}. Default
14274	value is 0.
14275	@end table
14276
14277	@subsection Examples
14278
14279	@itemize
14280	@item
14281	Use @command{ffmpeg} for a typical stabilization with default values:
14282	@example
14283	ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
14284	@end example
14285
14286	Note the use of the @ref{unsharp} filter which is always recommended.
14287
14288	@item
14289	Zoom in a bit more and load transform data from a given file:
14290	@example
14291	vidstabtransform=zoom=5:input="mytransforms.trf"
14292	@end example
14293
14294	@item
14295	Smoothen the video even more:
14296	@example
14297	vidstabtransform=smoothing=30
14298	@end example
14299	@end itemize
14300
14301	@section vflip
14302
14303	Flip the input video vertically.
14304
14305	For example, to vertically flip a video with @command{ffmpeg}:
14306	@example
14307	ffmpeg -i in.avi -vf "vflip" out.avi
14308	@end example
14309
14310	@anchor{vignette}
14311	@section vignette
14312
14313	Make or reverse a natural vignetting effect.
14314
14315	The filter accepts the following options:
14316
14317	@table @option
14318	@item angle, a
14319	Set lens angle expression as a number of radians.
14320
14321	The value is clipped in the @code{[0,PI/2]} range.
14322
14323	Default value: @code{"PI/5"}
14324
14325	@item x0
14326	@item y0
14327	Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
14328	by default.
14329
14330	@item mode
14331	Set forward/backward mode.
14332
14333	Available modes are:
14334	@table @samp
14335	@item forward
14336	The larger the distance from the central point, the darker the image becomes.
14337
14338	@item backward
14339	The larger the distance from the central point, the brighter the image becomes.
14340	This can be used to reverse a vignette effect, though there is no automatic
14341	detection to extract the lens @option{angle} and other settings (yet). It can
14342	also be used to create a burning effect.
14343	@end table
14344
14345	Default value is @samp{forward}.
14346
14347	@item eval
14348	Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
14349
14350	It accepts the following values:
14351	@table @samp
14352	@item init
14353	Evaluate expressions only once during the filter initialization.
14354
14355	@item frame
14356	Evaluate expressions for each incoming frame. This is way slower than the
14357	@samp{init} mode since it requires all the scalers to be re-computed, but it
14358	allows advanced dynamic expressions.
14359	@end table
14360
14361	Default value is @samp{init}.
14362
14363	@item dither
14364	Set dithering to reduce the circular banding effects. Default is @code{1}
14365	(enabled).
14366
14367	@item aspect
14368	Set vignette aspect. This setting allows one to adjust the shape of the vignette.
14369	Setting this value to the SAR of the input will make a rectangular vignetting
14370	following the dimensions of the video.
14371
14372	Default is @code{1/1}.
14373	@end table
14374
14375	@subsection Expressions
14376
14377	The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
14378	following parameters.
14379
14380	@table @option
14381	@item w
14382	@item h
14383	input width and height
14384
14385	@item n
14386	the number of input frame, starting from 0
14387
14388	@item pts
14389	the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
14390	@var{TB} units, NAN if undefined
14391
14392	@item r
14393	frame rate of the input video, NAN if the input frame rate is unknown
14394
14395	@item t
14396	the PTS (Presentation TimeStamp) of the filtered video frame,
14397	expressed in seconds, NAN if undefined
14398
14399	@item tb
14400	time base of the input video
14401	@end table
14402
14403
14404	@subsection Examples
14405
14406	@itemize
14407	@item
14408	Apply simple strong vignetting effect:
14409	@example
14410	vignette=PI/4
14411	@end example
14412
14413	@item
14414	Make a flickering vignetting:
14415	@example
14416	vignette='PI/4+random(1)*PI/50':eval=frame
14417	@end example
14418
14419	@end itemize
14420
14421	@section vstack
14422	Stack input videos vertically.
14423
14424	All streams must be of same pixel format and of same width.
14425
14426	Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
14427	to create same output.
14428
14429	The filter accept the following option:
14430
14431	@table @option
14432	@item inputs
14433	Set number of input streams. Default is 2.
14434
14435	@item shortest
14436	If set to 1, force the output to terminate when the shortest input
14437	terminates. Default value is 0.
14438	@end table
14439
14440	@section w3fdif
14441
14442	Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
14443	Deinterlacing Filter").
14444
14445	Based on the process described by Martin Weston for BBC R&D, and
14446	implemented based on the de-interlace algorithm written by Jim
14447	Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
14448	uses filter coefficients calculated by BBC R&D.
14449
14450	There are two sets of filter coefficients, so called "simple":
14451	and "complex". Which set of filter coefficients is used can
14452	be set by passing an optional parameter:
14453
14454	@table @option
14455	@item filter
14456	Set the interlacing filter coefficients. Accepts one of the following values:
14457
14458	@table @samp
14459	@item simple
14460	Simple filter coefficient set.
14461	@item complex
14462	More-complex filter coefficient set.
14463	@end table
14464	Default value is @samp{complex}.
14465
14466	@item deint
14467	Specify which frames to deinterlace. Accept one of the following values:
14468
14469	@table @samp
14470	@item all
14471	Deinterlace all frames,
14472	@item interlaced
14473	Only deinterlace frames marked as interlaced.
14474	@end table
14475
14476	Default value is @samp{all}.
14477	@end table
14478
14479	@section waveform
14480	Video waveform monitor.
14481
14482	The waveform monitor plots color component intensity. By default luminance
14483	only. Each column of the waveform corresponds to a column of pixels in the
14484	source video.
14485
14486	It accepts the following options:
14487
14488	@table @option
14489	@item mode, m
14490	Can be either @code{row}, or @code{column}. Default is @code{column}.
14491	In row mode, the graph on the left side represents color component value 0 and
14492	the right side represents value = 255. In column mode, the top side represents
14493	color component value = 0 and bottom side represents value = 255.
14494
14495	@item intensity, i
14496	Set intensity. Smaller values are useful to find out how many values of the same
14497	luminance are distributed across input rows/columns.
14498	Default value is @code{0.04}. Allowed range is [0, 1].
14499
14500	@item mirror, r
14501	Set mirroring mode. @code{0} means unmirrored, @code{1} means mirrored.
14502	In mirrored mode, higher values will be represented on the left
14503	side for @code{row} mode and at the top for @code{column} mode. Default is
14504	@code{1} (mirrored).
14505
14506	@item display, d
14507	Set display mode.
14508	It accepts the following values:
14509	@table @samp
14510	@item overlay
14511	Presents information identical to that in the @code{parade}, except
14512	that the graphs representing color components are superimposed directly
14513	over one another.
14514
14515	This display mode makes it easier to spot relative differences or similarities
14516	in overlapping areas of the color components that are supposed to be identical,
14517	such as neutral whites, grays, or blacks.
14518
14519	@item stack
14520	Display separate graph for the color components side by side in
14521	@code{row} mode or one below the other in @code{column} mode.
14522
14523	@item parade
14524	Display separate graph for the color components side by side in
14525	@code{column} mode or one below the other in @code{row} mode.
14526
14527	Using this display mode makes it easy to spot color casts in the highlights
14528	and shadows of an image, by comparing the contours of the top and the bottom
14529	graphs of each waveform. Since whites, grays, and blacks are characterized
14530	by exactly equal amounts of red, green, and blue, neutral areas of the picture
14531	should display three waveforms of roughly equal width/height. If not, the
14532	correction is easy to perform by making level adjustments the three waveforms.
14533	@end table
14534	Default is @code{stack}.
14535
14536	@item components, c
14537	Set which color components to display. Default is 1, which means only luminance
14538	or red color component if input is in RGB colorspace. If is set for example to
14539	7 it will display all 3 (if) available color components.
14540
14541	@item envelope, e
14542	@table @samp
14543	@item none
14544	No envelope, this is default.
14545
14546	@item instant
14547	Instant envelope, minimum and maximum values presented in graph will be easily
14548	visible even with small @code{step} value.
14549
14550	@item peak
14551	Hold minimum and maximum values presented in graph across time. This way you
14552	can still spot out of range values without constantly looking at waveforms.
14553
14554	@item peak+instant
14555	Peak and instant envelope combined together.
14556	@end table
14557
14558	@item filter, f
14559	@table @samp
14560	@item lowpass
14561	No filtering, this is default.
14562
14563	@item flat
14564	Luma and chroma combined together.
14565
14566	@item aflat
14567	Similar as above, but shows difference between blue and red chroma.
14568
14569	@item chroma
14570	Displays only chroma.
14571
14572	@item color
14573	Displays actual color value on waveform.
14574
14575	@item acolor
14576	Similar as above, but with luma showing frequency of chroma values.
14577	@end table
14578
14579	@item graticule, g
14580	Set which graticule to display.
14581
14582	@table @samp
14583	@item none
14584	Do not display graticule.
14585
14586	@item green
14587	Display green graticule showing legal broadcast ranges.
14588	@end table
14589
14590	@item opacity, o
14591	Set graticule opacity.
14592
14593	@item flags, fl
14594	Set graticule flags.
14595
14596	@table @samp
14597	@item numbers
14598	Draw numbers above lines. By default enabled.
14599
14600	@item dots
14601	Draw dots instead of lines.
14602	@end table
14603
14604	@item scale, s
14605	Set scale used for displaying graticule.
14606
14607	@table @samp
14608	@item digital
14609	@item millivolts
14610	@item ire
14611	@end table
14612	Default is digital.
14613
14614	@item bgopacity, b
14615	Set background opacity.
14616	@end table
14617
14618	@section weave
14619
14620	The @code{weave} takes a field-based video input and join
14621	each two sequential fields into single frame, producing a new double
14622	height clip with half the frame rate and half the frame count.
14623
14624	It accepts the following option:
14625
14626	@table @option
14627	@item first_field
14628	Set first field. Available values are:
14629
14630	@table @samp
14631	@item top, t
14632	Set the frame as top-field-first.
14633
14634	@item bottom, b
14635	Set the frame as bottom-field-first.
14636	@end table
14637	@end table
14638
14639	@subsection Examples
14640
14641	@itemize
14642	@item
14643	Interlace video using @ref{select} and @ref{separatefields} filter:
14644	@example
14645	separatefields,select=eq(mod(n,4),0)+eq(mod(n,4),3),weave
14646	@end example
14647	@end itemize
14648
14649	@section xbr
14650	Apply the xBR high-quality magnification filter which is designed for pixel
14651	art. It follows a set of edge-detection rules, see
14652	@url{http://www.libretro.com/forums/viewtopic.php?f=6&t=134}.
14653
14654	It accepts the following option:
14655
14656	@table @option
14657	@item n
14658	Set the scaling dimension: @code{2} for @code{2xBR}, @code{3} for
14659	@code{3xBR} and @code{4} for @code{4xBR}.
14660	Default is @code{3}.
14661	@end table
14662
14663	@anchor{yadif}
14664	@section yadif
14665
14666	Deinterlace the input video ("yadif" means "yet another deinterlacing
14667	filter").
14668
14669	It accepts the following parameters:
14670
14671
14672	@table @option
14673
14674	@item mode
14675	The interlacing mode to adopt. It accepts one of the following values:
14676
14677	@table @option
14678	@item 0, send_frame
14679	Output one frame for each frame.
14680	@item 1, send_field
14681	Output one frame for each field.
14682	@item 2, send_frame_nospatial
14683	Like @code{send_frame}, but it skips the spatial interlacing check.
14684	@item 3, send_field_nospatial
14685	Like @code{send_field}, but it skips the spatial interlacing check.
14686	@end table
14687
14688	The default value is @code{send_frame}.
14689
14690	@item parity
14691	The picture field parity assumed for the input interlaced video. It accepts one
14692	of the following values:
14693
14694	@table @option
14695	@item 0, tff
14696	Assume the top field is first.
14697	@item 1, bff
14698	Assume the bottom field is first.
14699	@item -1, auto
14700	Enable automatic detection of field parity.
14701	@end table
14702
14703	The default value is @code{auto}.
14704	If the interlacing is unknown or the decoder does not export this information,
14705	top field first will be assumed.
14706
14707	@item deint
14708	Specify which frames to deinterlace. Accept one of the following
14709	values:
14710
14711	@table @option
14712	@item 0, all
14713	Deinterlace all frames.
14714	@item 1, interlaced
14715	Only deinterlace frames marked as interlaced.
14716	@end table
14717
14718	The default value is @code{all}.
14719	@end table
14720
14721	@section zoompan
14722
14723	Apply Zoom & Pan effect.
14724
14725	This filter accepts the following options:
14726
14727	@table @option
14728	@item zoom, z
14729	Set the zoom expression. Default is 1.
14730
14731	@item x
14732	@item y
14733	Set the x and y expression. Default is 0.
14734
14735	@item d
14736	Set the duration expression in number of frames.
14737	This sets for how many number of frames effect will last for
14738	single input image.
14739
14740	@item s
14741	Set the output image size, default is 'hd720'.
14742
14743	@item fps
14744	Set the output frame rate, default is '25'.
14745	@end table
14746
14747	Each expression can contain the following constants:
14748
14749	@table @option
14750	@item in_w, iw
14751	Input width.
14752
14753	@item in_h, ih
14754	Input height.
14755
14756	@item out_w, ow
14757	Output width.
14758
14759	@item out_h, oh
14760	Output height.
14761
14762	@item in
14763	Input frame count.
14764
14765	@item on
14766	Output frame count.
14767
14768	@item x
14769	@item y
14770	Last calculated 'x' and 'y' position from 'x' and 'y' expression
14771	for current input frame.
14772
14773	@item px
14774	@item py
14775	'x' and 'y' of last output frame of previous input frame or 0 when there was
14776	not yet such frame (first input frame).
14777
14778	@item zoom
14779	Last calculated zoom from 'z' expression for current input frame.
14780
14781	@item pzoom
14782	Last calculated zoom of last output frame of previous input frame.
14783
14784	@item duration
14785	Number of output frames for current input frame. Calculated from 'd' expression
14786	for each input frame.
14787
14788	@item pduration
14789	number of output frames created for previous input frame
14790
14791	@item a
14792	Rational number: input width / input height
14793
14794	@item sar
14795	sample aspect ratio
14796
14797	@item dar
14798	display aspect ratio
14799
14800	@end table
14801
14802	@subsection Examples
14803
14804	@itemize
14805	@item
14806	Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
14807	@example
14808	zoompan=z='min(zoom+0.0015,1.5)':d=700:x='if(gte(zoom,1.5),x,x+1/a)':y='if(gte(zoom,1.5),y,y+1)':s=640x360
14809	@end example
14810
14811	@item
14812	Zoom-in up to 1.5 and pan always at center of picture:
14813	@example
14814	zoompan=z='min(zoom+0.0015,1.5)':d=700:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
14815	@end example
14816
14817	@item
14818	Same as above but without pausing:
14819	@example
14820	zoompan=z='min(max(zoom,pzoom)+0.0015,1.5)':d=1:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
14821	@end example
14822	@end itemize
14823
14824	@section zscale
14825	Scale (resize) the input video, using the z.lib library:
14826	https://github.com/sekrit-twc/zimg.
14827
14828	The zscale filter forces the output display aspect ratio to be the same
14829	as the input, by changing the output sample aspect ratio.
14830
14831	If the input image format is different from the format requested by
14832	the next filter, the zscale filter will convert the input to the
14833	requested format.
14834
14835	@subsection Options
14836	The filter accepts the following options.
14837
14838	@table @option
14839	@item width, w
14840	@item height, h
14841	Set the output video dimension expression. Default value is the input
14842	dimension.
14843
14844	If the @var{width} or @var{w} is 0, the input width is used for the output.
14845	If the @var{height} or @var{h} is 0, the input height is used for the output.
14846
14847	If one of the values is -1, the zscale filter will use a value that
14848	maintains the aspect ratio of the input image, calculated from the
14849	other specified dimension. If both of them are -1, the input size is
14850	used
14851
14852	If one of the values is -n with n > 1, the zscale filter will also use a value
14853	that maintains the aspect ratio of the input image, calculated from the other
14854	specified dimension. After that it will, however, make sure that the calculated
14855	dimension is divisible by n and adjust the value if necessary.
14856
14857	See below for the list of accepted constants for use in the dimension
14858	expression.
14859
14860	@item size, s
14861	Set the video size. For the syntax of this option, check the
14862	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14863
14864	@item dither, d
14865	Set the dither type.
14866
14867	Possible values are:
14868	@table @var
14869	@item none
14870	@item ordered
14871	@item random
14872	@item error_diffusion
14873	@end table
14874
14875	Default is none.
14876
14877	@item filter, f
14878	Set the resize filter type.
14879
14880	Possible values are:
14881	@table @var
14882	@item point
14883	@item bilinear
14884	@item bicubic
14885	@item spline16
14886	@item spline36
14887	@item lanczos
14888	@end table
14889
14890	Default is bilinear.
14891
14892	@item range, r
14893	Set the color range.
14894
14895	Possible values are:
14896	@table @var
14897	@item input
14898	@item limited
14899	@item full
14900	@end table
14901
14902	Default is same as input.
14903
14904	@item primaries, p
14905	Set the color primaries.
14906
14907	Possible values are:
14908	@table @var
14909	@item input
14910	@item 709
14911	@item unspecified
14912	@item 170m
14913	@item 240m
14914	@item 2020
14915	@end table
14916
14917	Default is same as input.
14918
14919	@item transfer, t
14920	Set the transfer characteristics.
14921
14922	Possible values are:
14923	@table @var
14924	@item input
14925	@item 709
14926	@item unspecified
14927	@item 601
14928	@item linear
14929	@item 2020_10
14930	@item 2020_12
14931	@item smpte2084
14932	@item iec61966-2-1
14933	@item arib-std-b67
14934	@end table
14935
14936	Default is same as input.
14937
14938	@item matrix, m
14939	Set the colorspace matrix.
14940
14941	Possible value are:
14942	@table @var
14943	@item input
14944	@item 709
14945	@item unspecified
14946	@item 470bg
14947	@item 170m
14948	@item 2020_ncl
14949	@item 2020_cl
14950	@end table
14951
14952	Default is same as input.
14953
14954	@item rangein, rin
14955	Set the input color range.
14956
14957	Possible values are:
14958	@table @var
14959	@item input
14960	@item limited
14961	@item full
14962	@end table
14963
14964	Default is same as input.
14965
14966	@item primariesin, pin
14967	Set the input color primaries.
14968
14969	Possible values are:
14970	@table @var
14971	@item input
14972	@item 709
14973	@item unspecified
14974	@item 170m
14975	@item 240m
14976	@item 2020
14977	@end table
14978
14979	Default is same as input.
14980
14981	@item transferin, tin
14982	Set the input transfer characteristics.
14983
14984	Possible values are:
14985	@table @var
14986	@item input
14987	@item 709
14988	@item unspecified
14989	@item 601
14990	@item linear
14991	@item 2020_10
14992	@item 2020_12
14993	@end table
14994
14995	Default is same as input.
14996
14997	@item matrixin, min
14998	Set the input colorspace matrix.
14999
15000	Possible value are:
15001	@table @var
15002	@item input
15003	@item 709
15004	@item unspecified
15005	@item 470bg
15006	@item 170m
15007	@item 2020_ncl
15008	@item 2020_cl
15009	@end table
15010
15011	@item chromal, c
15012	Set the output chroma location.
15013
15014	Possible values are:
15015	@table @var
15016	@item input
15017	@item left
15018	@item center
15019	@item topleft
15020	@item top
15021	@item bottomleft
15022	@item bottom
15023	@end table
15024
15025	@item chromalin, cin
15026	Set the input chroma location.
15027
15028	Possible values are:
15029	@table @var
15030	@item input
15031	@item left
15032	@item center
15033	@item topleft
15034	@item top
15035	@item bottomleft
15036	@item bottom
15037	@end table
15038
15039	@item npl
15040	Set the nominal peak luminance.
15041	@end table
15042
15043	The values of the @option{w} and @option{h} options are expressions
15044	containing the following constants:
15045
15046	@table @var
15047	@item in_w
15048	@item in_h
15049	The input width and height
15050
15051	@item iw
15052	@item ih
15053	These are the same as @var{in_w} and @var{in_h}.
15054
15055	@item out_w
15056	@item out_h
15057	The output (scaled) width and height
15058
15059	@item ow
15060	@item oh
15061	These are the same as @var{out_w} and @var{out_h}
15062
15063	@item a
15064	The same as @var{iw} / @var{ih}
15065
15066	@item sar
15067	input sample aspect ratio
15068
15069	@item dar
15070	The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
15071
15072	@item hsub
15073	@item vsub
15074	horizontal and vertical input chroma subsample values. For example for the
15075	pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
15076
15077	@item ohsub
15078	@item ovsub
15079	horizontal and vertical output chroma subsample values. For example for the
15080	pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
15081	@end table
15082
15083	@table @option
15084	@end table
15085
15086	@c man end VIDEO FILTERS
15087
15088	@chapter Video Sources
15089	@c man begin VIDEO SOURCES
15090
15091	Below is a description of the currently available video sources.
15092
15093	@section buffer
15094
15095	Buffer video frames, and make them available to the filter chain.
15096
15097	This source is mainly intended for a programmatic use, in particular
15098	through the interface defined in @file{libavfilter/vsrc_buffer.h}.
15099
15100	It accepts the following parameters:
15101
15102	@table @option
15103
15104	@item video_size
15105	Specify the size (width and height) of the buffered video frames. For the
15106	syntax of this option, check the
15107	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15108
15109	@item width
15110	The input video width.
15111
15112	@item height
15113	The input video height.
15114
15115	@item pix_fmt
15116	A string representing the pixel format of the buffered video frames.
15117	It may be a number corresponding to a pixel format, or a pixel format
15118	name.
15119
15120	@item time_base
15121	Specify the timebase assumed by the timestamps of the buffered frames.
15122
15123	@item frame_rate
15124	Specify the frame rate expected for the video stream.
15125
15126	@item pixel_aspect, sar
15127	The sample (pixel) aspect ratio of the input video.
15128
15129	@item sws_param
15130	Specify the optional parameters to be used for the scale filter which
15131	is automatically inserted when an input change is detected in the
15132	input size or format.
15133
15134	@item hw_frames_ctx
15135	When using a hardware pixel format, this should be a reference to an
15136	AVHWFramesContext describing input frames.
15137	@end table
15138
15139	For example:
15140	@example
15141	buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
15142	@end example
15143
15144	will instruct the source to accept video frames with size 320x240 and
15145	with format "yuv410p", assuming 1/24 as the timestamps timebase and
15146	square pixels (1:1 sample aspect ratio).
15147	Since the pixel format with name "yuv410p" corresponds to the number 6
15148	(check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
15149	this example corresponds to:
15150	@example
15151	buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
15152	@end example
15153
15154	Alternatively, the options can be specified as a flat string, but this
15155	syntax is deprecated:
15156
15157	@var{width}:@var{height}:@var{pix_fmt}:@var{time_base.num}:@var{time_base.den}:@var{pixel_aspect.num}:@var{pixel_aspect.den}[:@var{sws_param}]
15158
15159	@section cellauto
15160
15161	Create a pattern generated by an elementary cellular automaton.
15162
15163	The initial state of the cellular automaton can be defined through the
15164	@option{filename} and @option{pattern} options. If such options are
15165	not specified an initial state is created randomly.
15166
15167	At each new frame a new row in the video is filled with the result of
15168	the cellular automaton next generation. The behavior when the whole
15169	frame is filled is defined by the @option{scroll} option.
15170
15171	This source accepts the following options:
15172
15173	@table @option
15174	@item filename, f
15175	Read the initial cellular automaton state, i.e. the starting row, from
15176	the specified file.
15177	In the file, each non-whitespace character is considered an alive
15178	cell, a newline will terminate the row, and further characters in the
15179	file will be ignored.
15180
15181	@item pattern, p
15182	Read the initial cellular automaton state, i.e. the starting row, from
15183	the specified string.
15184
15185	Each non-whitespace character in the string is considered an alive
15186	cell, a newline will terminate the row, and further characters in the
15187	string will be ignored.
15188
15189	@item rate, r
15190	Set the video rate, that is the number of frames generated per second.
15191	Default is 25.
15192
15193	@item random_fill_ratio, ratio
15194	Set the random fill ratio for the initial cellular automaton row. It
15195	is a floating point number value ranging from 0 to 1, defaults to
15196	1/PHI.
15197
15198	This option is ignored when a file or a pattern is specified.
15199
15200	@item random_seed, seed
15201	Set the seed for filling randomly the initial row, must be an integer
15202	included between 0 and UINT32_MAX. If not specified, or if explicitly
15203	set to -1, the filter will try to use a good random seed on a best
15204	effort basis.
15205
15206	@item rule
15207	Set the cellular automaton rule, it is a number ranging from 0 to 255.
15208	Default value is 110.
15209
15210	@item size, s
15211	Set the size of the output video. For the syntax of this option, check the
15212	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15213
15214	If @option{filename} or @option{pattern} is specified, the size is set
15215	by default to the width of the specified initial state row, and the
15216	height is set to @var{width} * PHI.
15217
15218	If @option{size} is set, it must contain the width of the specified
15219	pattern string, and the specified pattern will be centered in the
15220	larger row.
15221
15222	If a filename or a pattern string is not specified, the size value
15223	defaults to "320x518" (used for a randomly generated initial state).
15224
15225	@item scroll
15226	If set to 1, scroll the output upward when all the rows in the output
15227	have been already filled. If set to 0, the new generated row will be
15228	written over the top row just after the bottom row is filled.
15229	Defaults to 1.
15230
15231	@item start_full, full
15232	If set to 1, completely fill the output with generated rows before
15233	outputting the first frame.
15234	This is the default behavior, for disabling set the value to 0.
15235
15236	@item stitch
15237	If set to 1, stitch the left and right row edges together.
15238	This is the default behavior, for disabling set the value to 0.
15239	@end table
15240
15241	@subsection Examples
15242
15243	@itemize
15244	@item
15245	Read the initial state from @file{pattern}, and specify an output of
15246	size 200x400.
15247	@example
15248	cellauto=f=pattern:s=200x400
15249	@end example
15250
15251	@item
15252	Generate a random initial row with a width of 200 cells, with a fill
15253	ratio of 2/3:
15254	@example
15255	cellauto=ratio=2/3:s=200x200
15256	@end example
15257
15258	@item
15259	Create a pattern generated by rule 18 starting by a single alive cell
15260	centered on an initial row with width 100:
15261	@example
15262	cellauto=p=@@:s=100x400:full=0:rule=18
15263	@end example
15264
15265	@item
15266	Specify a more elaborated initial pattern:
15267	@example
15268	cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
15269	@end example
15270
15271	@end itemize
15272
15273	@anchor{coreimagesrc}
15274	@section coreimagesrc
15275	Video source generated on GPU using Apple's CoreImage API on OSX.
15276
15277	This video source is a specialized version of the @ref{coreimage} video filter.
15278	Use a core image generator at the beginning of the applied filterchain to
15279	generate the content.
15280
15281	The coreimagesrc video source accepts the following options:
15282	@table @option
15283	@item list_generators
15284	List all available generators along with all their respective options as well as
15285	possible minimum and maximum values along with the default values.
15286	@example
15287	list_generators=true
15288	@end example
15289
15290	@item size, s
15291	Specify the size of the sourced video. For the syntax of this option, check the
15292	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15293	The default value is @code{320x240}.
15294
15295	@item rate, r
15296	Specify the frame rate of the sourced video, as the number of frames
15297	generated per second. It has to be a string in the format
15298	@var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15299	number or a valid video frame rate abbreviation. The default value is
15300	"25".
15301
15302	@item sar
15303	Set the sample aspect ratio of the sourced video.
15304
15305	@item duration, d
15306	Set the duration of the sourced video. See
15307	@ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15308	for the accepted syntax.
15309
15310	If not specified, or the expressed duration is negative, the video is
15311	supposed to be generated forever.
15312	@end table
15313
15314	Additionally, all options of the @ref{coreimage} video filter are accepted.
15315	A complete filterchain can be used for further processing of the
15316	generated input without CPU-HOST transfer. See @ref{coreimage} documentation
15317	and examples for details.
15318
15319	@subsection Examples
15320
15321	@itemize
15322
15323	@item
15324	Use CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
15325	given as complete and escaped command-line for Apple's standard bash shell:
15326	@example
15327	ffmpeg -f lavfi -i coreimagesrc=s=100x100:filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
15328	@end example
15329	This example is equivalent to the QRCode example of @ref{coreimage} without the
15330	need for a nullsrc video source.
15331	@end itemize
15332
15333
15334	@section mandelbrot
15335
15336	Generate a Mandelbrot set fractal, and progressively zoom towards the
15337	point specified with @var{start_x} and @var{start_y}.
15338
15339	This source accepts the following options:
15340
15341	@table @option
15342
15343	@item end_pts
15344	Set the terminal pts value. Default value is 400.
15345
15346	@item end_scale
15347	Set the terminal scale value.
15348	Must be a floating point value. Default value is 0.3.
15349
15350	@item inner
15351	Set the inner coloring mode, that is the algorithm used to draw the
15352	Mandelbrot fractal internal region.
15353
15354	It shall assume one of the following values:
15355	@table @option
15356	@item black
15357	Set black mode.
15358	@item convergence
15359	Show time until convergence.
15360	@item mincol
15361	Set color based on point closest to the origin of the iterations.
15362	@item period
15363	Set period mode.
15364	@end table
15365
15366	Default value is @var{mincol}.
15367
15368	@item bailout
15369	Set the bailout value. Default value is 10.0.
15370
15371	@item maxiter
15372	Set the maximum of iterations performed by the rendering
15373	algorithm. Default value is 7189.
15374
15375	@item outer
15376	Set outer coloring mode.
15377	It shall assume one of following values:
15378	@table @option
15379	@item iteration_count
15380	Set iteration cound mode.
15381	@item normalized_iteration_count
15382	set normalized iteration count mode.
15383	@end table
15384	Default value is @var{normalized_iteration_count}.
15385
15386	@item rate, r
15387	Set frame rate, expressed as number of frames per second. Default
15388	value is "25".
15389
15390	@item size, s
15391	Set frame size. For the syntax of this option, check the "Video
15392	size" section in the ffmpeg-utils manual. Default value is "640x480".
15393
15394	@item start_scale
15395	Set the initial scale value. Default value is 3.0.
15396
15397	@item start_x
15398	Set the initial x position. Must be a floating point value between
15399	-100 and 100. Default value is -0.743643887037158704752191506114774.
15400
15401	@item start_y
15402	Set the initial y position. Must be a floating point value between
15403	-100 and 100. Default value is -0.131825904205311970493132056385139.
15404	@end table
15405
15406	@section mptestsrc
15407
15408	Generate various test patterns, as generated by the MPlayer test filter.
15409
15410	The size of the generated video is fixed, and is 256x256.
15411	This source is useful in particular for testing encoding features.
15412
15413	This source accepts the following options:
15414
15415	@table @option
15416
15417	@item rate, r
15418	Specify the frame rate of the sourced video, as the number of frames
15419	generated per second. It has to be a string in the format
15420	@var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15421	number or a valid video frame rate abbreviation. The default value is
15422	"25".
15423
15424	@item duration, d
15425	Set the duration of the sourced video. See
15426	@ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15427	for the accepted syntax.
15428
15429	If not specified, or the expressed duration is negative, the video is
15430	supposed to be generated forever.
15431
15432	@item test, t
15433
15434	Set the number or the name of the test to perform. Supported tests are:
15435	@table @option
15436	@item dc_luma
15437	@item dc_chroma
15438	@item freq_luma
15439	@item freq_chroma
15440	@item amp_luma
15441	@item amp_chroma
15442	@item cbp
15443	@item mv
15444	@item ring1
15445	@item ring2
15446	@item all
15447
15448	@end table
15449
15450	Default value is "all", which will cycle through the list of all tests.
15451	@end table
15452
15453	Some examples:
15454	@example
15455	mptestsrc=t=dc_luma
15456	@end example
15457
15458	will generate a "dc_luma" test pattern.
15459
15460	@section frei0r_src
15461
15462	Provide a frei0r source.
15463
15464	To enable compilation of this filter you need to install the frei0r
15465	header and configure FFmpeg with @code{--enable-frei0r}.
15466
15467	This source accepts the following parameters:
15468
15469	@table @option
15470
15471	@item size
15472	The size of the video to generate. For the syntax of this option, check the
15473	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15474
15475	@item framerate
15476	The framerate of the generated video. It may be a string of the form
15477	@var{num}/@var{den} or a frame rate abbreviation.
15478
15479	@item filter_name
15480	The name to the frei0r source to load. For more information regarding frei0r and
15481	how to set the parameters, read the @ref{frei0r} section in the video filters
15482	documentation.
15483
15484	@item filter_params
15485	A '|'-separated list of parameters to pass to the frei0r source.
15486
15487	@end table
15488
15489	For example, to generate a frei0r partik0l source with size 200x200
15490	and frame rate 10 which is overlaid on the overlay filter main input:
15491	@example
15492	frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
15493	@end example
15494
15495	@section life
15496
15497	Generate a life pattern.
15498
15499	This source is based on a generalization of John Conway's life game.
15500
15501	The sourced input represents a life grid, each pixel represents a cell
15502	which can be in one of two possible states, alive or dead. Every cell
15503	interacts with its eight neighbours, which are the cells that are
15504	horizontally, vertically, or diagonally adjacent.
15505
15506	At each interaction the grid evolves according to the adopted rule,
15507	which specifies the number of neighbor alive cells which will make a
15508	cell stay alive or born. The @option{rule} option allows one to specify
15509	the rule to adopt.
15510
15511	This source accepts the following options:
15512
15513	@table @option
15514	@item filename, f
15515	Set the file from which to read the initial grid state. In the file,
15516	each non-whitespace character is considered an alive cell, and newline
15517	is used to delimit the end of each row.
15518
15519	If this option is not specified, the initial grid is generated
15520	randomly.
15521
15522	@item rate, r
15523	Set the video rate, that is the number of frames generated per second.
15524	Default is 25.
15525
15526	@item random_fill_ratio, ratio
15527	Set the random fill ratio for the initial random grid. It is a
15528	floating point number value ranging from 0 to 1, defaults to 1/PHI.
15529	It is ignored when a file is specified.
15530
15531	@item random_seed, seed
15532	Set the seed for filling the initial random grid, must be an integer
15533	included between 0 and UINT32_MAX. If not specified, or if explicitly
15534	set to -1, the filter will try to use a good random seed on a best
15535	effort basis.
15536
15537	@item rule
15538	Set the life rule.
15539
15540	A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
15541	where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
15542	@var{NS} specifies the number of alive neighbor cells which make a
15543	live cell stay alive, and @var{NB} the number of alive neighbor cells
15544	which make a dead cell to become alive (i.e. to "born").
15545	"s" and "b" can be used in place of "S" and "B", respectively.
15546
15547	Alternatively a rule can be specified by an 18-bits integer. The 9
15548	high order bits are used to encode the next cell state if it is alive
15549	for each number of neighbor alive cells, the low order bits specify
15550	the rule for "borning" new cells. Higher order bits encode for an
15551	higher number of neighbor cells.
15552	For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
15553	rule of 12 and a born rule of 9, which corresponds to "S23/B03".
15554
15555	Default value is "S23/B3", which is the original Conway's game of life
15556	rule, and will keep a cell alive if it has 2 or 3 neighbor alive
15557	cells, and will born a new cell if there are three alive cells around
15558	a dead cell.
15559
15560	@item size, s
15561	Set the size of the output video. For the syntax of this option, check the
15562	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15563
15564	If @option{filename} is specified, the size is set by default to the
15565	same size of the input file. If @option{size} is set, it must contain
15566	the size specified in the input file, and the initial grid defined in
15567	that file is centered in the larger resulting area.
15568
15569	If a filename is not specified, the size value defaults to "320x240"
15570	(used for a randomly generated initial grid).
15571
15572	@item stitch
15573	If set to 1, stitch the left and right grid edges together, and the
15574	top and bottom edges also. Defaults to 1.
15575
15576	@item mold
15577	Set cell mold speed. If set, a dead cell will go from @option{death_color} to
15578	@option{mold_color} with a step of @option{mold}. @option{mold} can have a
15579	value from 0 to 255.
15580
15581	@item life_color
15582	Set the color of living (or new born) cells.
15583
15584	@item death_color
15585	Set the color of dead cells. If @option{mold} is set, this is the first color
15586	used to represent a dead cell.
15587
15588	@item mold_color
15589	Set mold color, for definitely dead and moldy cells.
15590
15591	For the syntax of these 3 color options, check the "Color" section in the
15592	ffmpeg-utils manual.
15593	@end table
15594
15595	@subsection Examples
15596
15597	@itemize
15598	@item
15599	Read a grid from @file{pattern}, and center it on a grid of size
15600	300x300 pixels:
15601	@example
15602	life=f=pattern:s=300x300
15603	@end example
15604
15605	@item
15606	Generate a random grid of size 200x200, with a fill ratio of 2/3:
15607	@example
15608	life=ratio=2/3:s=200x200
15609	@end example
15610
15611	@item
15612	Specify a custom rule for evolving a randomly generated grid:
15613	@example
15614	life=rule=S14/B34
15615	@end example
15616
15617	@item
15618	Full example with slow death effect (mold) using @command{ffplay}:
15619	@example
15620	ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
15621	@end example
15622	@end itemize
15623
15624	@anchor{allrgb}
15625	@anchor{allyuv}
15626	@anchor{color}
15627	@anchor{haldclutsrc}
15628	@anchor{nullsrc}
15629	@anchor{rgbtestsrc}
15630	@anchor{smptebars}
15631	@anchor{smptehdbars}
15632	@anchor{testsrc}
15633	@anchor{testsrc2}
15634	@anchor{yuvtestsrc}
15635	@section allrgb, allyuv, color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc, testsrc2, yuvtestsrc
15636
15637	The @code{allrgb} source returns frames of size 4096x4096 of all rgb colors.
15638
15639	The @code{allyuv} source returns frames of size 4096x4096 of all yuv colors.
15640
15641	The @code{color} source provides an uniformly colored input.
15642
15643	The @code{haldclutsrc} source provides an identity Hald CLUT. See also
15644	@ref{haldclut} filter.
15645
15646	The @code{nullsrc} source returns unprocessed video frames. It is
15647	mainly useful to be employed in analysis / debugging tools, or as the
15648	source for filters which ignore the input data.
15649
15650	The @code{rgbtestsrc} source generates an RGB test pattern useful for
15651	detecting RGB vs BGR issues. You should see a red, green and blue
15652	stripe from top to bottom.
15653
15654	The @code{smptebars} source generates a color bars pattern, based on
15655	the SMPTE Engineering Guideline EG 1-1990.
15656
15657	The @code{smptehdbars} source generates a color bars pattern, based on
15658	the SMPTE RP 219-2002.
15659
15660	The @code{testsrc} source generates a test video pattern, showing a
15661	color pattern, a scrolling gradient and a timestamp. This is mainly
15662	intended for testing purposes.
15663
15664	The @code{testsrc2} source is similar to testsrc, but supports more
15665	pixel formats instead of just @code{rgb24}. This allows using it as an
15666	input for other tests without requiring a format conversion.
15667
15668	The @code{yuvtestsrc} source generates an YUV test pattern. You should
15669	see a y, cb and cr stripe from top to bottom.
15670
15671	The sources accept the following parameters:
15672
15673	@table @option
15674
15675	@item color, c
15676	Specify the color of the source, only available in the @code{color}
15677	source. For the syntax of this option, check the "Color" section in the
15678	ffmpeg-utils manual.
15679
15680	@item level
15681	Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
15682	source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
15683	pixels to be used as identity matrix for 3D lookup tables. Each component is
15684	coded on a @code{1/(N*N)} scale.
15685
15686	@item size, s
15687	Specify the size of the sourced video. For the syntax of this option, check the
15688	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15689	The default value is @code{320x240}.
15690
15691	This option is not available with the @code{haldclutsrc} filter.
15692
15693	@item rate, r
15694	Specify the frame rate of the sourced video, as the number of frames
15695	generated per second. It has to be a string in the format
15696	@var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15697	number or a valid video frame rate abbreviation. The default value is
15698	"25".
15699
15700	@item sar
15701	Set the sample aspect ratio of the sourced video.
15702
15703	@item duration, d
15704	Set the duration of the sourced video. See
15705	@ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15706	for the accepted syntax.
15707
15708	If not specified, or the expressed duration is negative, the video is
15709	supposed to be generated forever.
15710
15711	@item decimals, n
15712	Set the number of decimals to show in the timestamp, only available in the
15713	@code{testsrc} source.
15714
15715	The displayed timestamp value will correspond to the original
15716	timestamp value multiplied by the power of 10 of the specified
15717	value. Default value is 0.
15718	@end table
15719
15720	For example the following:
15721	@example
15722	testsrc=duration=5.3:size=qcif:rate=10
15723	@end example
15724
15725	will generate a video with a duration of 5.3 seconds, with size
15726	176x144 and a frame rate of 10 frames per second.
15727
15728	The following graph description will generate a red source
15729	with an opacity of 0.2, with size "qcif" and a frame rate of 10
15730	frames per second.
15731	@example
15732	color=c=red@@0.2:s=qcif:r=10
15733	@end example
15734
15735	If the input content is to be ignored, @code{nullsrc} can be used. The
15736	following command generates noise in the luminance plane by employing
15737	the @code{geq} filter:
15738	@example
15739	nullsrc=s=256x256, geq=random(1)*255:128:128
15740	@end example
15741
15742	@subsection Commands
15743
15744	The @code{color} source supports the following commands:
15745
15746	@table @option
15747	@item c, color
15748	Set the color of the created image. Accepts the same syntax of the
15749	corresponding @option{color} option.
15750	@end table
15751
15752	@c man end VIDEO SOURCES
15753
15754	@chapter Video Sinks
15755	@c man begin VIDEO SINKS
15756
15757	Below is a description of the currently available video sinks.
15758
15759	@section buffersink
15760
15761	Buffer video frames, and make them available to the end of the filter
15762	graph.
15763
15764	This sink is mainly intended for programmatic use, in particular
15765	through the interface defined in @file{libavfilter/buffersink.h}
15766	or the options system.
15767
15768	It accepts a pointer to an AVBufferSinkContext structure, which
15769	defines the incoming buffers' formats, to be passed as the opaque
15770	parameter to @code{avfilter_init_filter} for initialization.
15771
15772	@section nullsink
15773
15774	Null video sink: do absolutely nothing with the input video. It is
15775	mainly useful as a template and for use in analysis / debugging
15776	tools.
15777
15778	@c man end VIDEO SINKS
15779
15780	@chapter Multimedia Filters
15781	@c man begin MULTIMEDIA FILTERS
15782
15783	Below is a description of the currently available multimedia filters.
15784
15785	@section abitscope
15786
15787	Convert input audio to a video output, displaying the audio bit scope.
15788
15789	The filter accepts the following options:
15790
15791	@table @option
15792	@item rate, r
15793	Set frame rate, expressed as number of frames per second. Default
15794	value is "25".
15795
15796	@item size, s
15797	Specify the video size for the output. For the syntax of this option, check the
15798	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15799	Default value is @code{1024x256}.
15800
15801	@item colors
15802	Specify list of colors separated by space or by '|' which will be used to
15803	draw channels. Unrecognized or missing colors will be replaced
15804	by white color.
15805	@end table
15806
15807	@section ahistogram
15808
15809	Convert input audio to a video output, displaying the volume histogram.
15810
15811	The filter accepts the following options:
15812
15813	@table @option
15814	@item dmode
15815	Specify how histogram is calculated.
15816
15817	It accepts the following values:
15818	@table @samp
15819	@item single
15820	Use single histogram for all channels.
15821	@item separate
15822	Use separate histogram for each channel.
15823	@end table
15824	Default is @code{single}.
15825
15826	@item rate, r
15827	Set frame rate, expressed as number of frames per second. Default
15828	value is "25".
15829
15830	@item size, s
15831	Specify the video size for the output. For the syntax of this option, check the
15832	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15833	Default value is @code{hd720}.
15834
15835	@item scale
15836	Set display scale.
15837
15838	It accepts the following values:
15839	@table @samp
15840	@item log
15841	logarithmic
15842	@item sqrt
15843	square root
15844	@item cbrt
15845	cubic root
15846	@item lin
15847	linear
15848	@item rlog
15849	reverse logarithmic
15850	@end table
15851	Default is @code{log}.
15852
15853	@item ascale
15854	Set amplitude scale.
15855
15856	It accepts the following values:
15857	@table @samp
15858	@item log
15859	logarithmic
15860	@item lin
15861	linear
15862	@end table
15863	Default is @code{log}.
15864
15865	@item acount
15866	Set how much frames to accumulate in histogram.
15867	Defauls is 1. Setting this to -1 accumulates all frames.
15868
15869	@item rheight
15870	Set histogram ratio of window height.
15871
15872	@item slide
15873	Set sonogram sliding.
15874
15875	It accepts the following values:
15876	@table @samp
15877	@item replace
15878	replace old rows with new ones.
15879	@item scroll
15880	scroll from top to bottom.
15881	@end table
15882	Default is @code{replace}.
15883	@end table
15884
15885	@section aphasemeter
15886
15887	Convert input audio to a video output, displaying the audio phase.
15888
15889	The filter accepts the following options:
15890
15891	@table @option
15892	@item rate, r
15893	Set the output frame rate. Default value is @code{25}.
15894
15895	@item size, s
15896	Set the video size for the output. For the syntax of this option, check the
15897	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15898	Default value is @code{800x400}.
15899
15900	@item rc
15901	@item gc
15902	@item bc
15903	Specify the red, green, blue contrast. Default values are @code{2},
15904	@code{7} and @code{1}.
15905	Allowed range is @code{[0, 255]}.
15906
15907	@item mpc
15908	Set color which will be used for drawing median phase. If color is
15909	@code{none} which is default, no median phase value will be drawn.
15910
15911	@item video
15912	Enable video output. Default is enabled.
15913	@end table
15914
15915	The filter also exports the frame metadata @code{lavfi.aphasemeter.phase} which
15916	represents mean phase of current audio frame. Value is in range @code{[-1, 1]}.
15917	The @code{-1} means left and right channels are completely out of phase and
15918	@code{1} means channels are in phase.
15919
15920	@section avectorscope
15921
15922	Convert input audio to a video output, representing the audio vector
15923	scope.
15924
15925	The filter is used to measure the difference between channels of stereo
15926	audio stream. A monoaural signal, consisting of identical left and right
15927	signal, results in straight vertical line. Any stereo separation is visible
15928	as a deviation from this line, creating a Lissajous figure.
15929	If the straight (or deviation from it) but horizontal line appears this
15930	indicates that the left and right channels are out of phase.
15931
15932	The filter accepts the following options:
15933
15934	@table @option
15935	@item mode, m
15936	Set the vectorscope mode.
15937
15938	Available values are:
15939	@table @samp
15940	@item lissajous
15941	Lissajous rotated by 45 degrees.
15942
15943	@item lissajous_xy
15944	Same as above but not rotated.
15945
15946	@item polar
15947	Shape resembling half of circle.
15948	@end table
15949
15950	Default value is @samp{lissajous}.
15951
15952	@item size, s
15953	Set the video size for the output. For the syntax of this option, check the
15954	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15955	Default value is @code{400x400}.
15956
15957	@item rate, r
15958	Set the output frame rate. Default value is @code{25}.
15959
15960	@item rc
15961	@item gc
15962	@item bc
15963	@item ac
15964	Specify the red, green, blue and alpha contrast. Default values are @code{40},
15965	@code{160}, @code{80} and @code{255}.
15966	Allowed range is @code{[0, 255]}.
15967
15968	@item rf
15969	@item gf
15970	@item bf
15971	@item af
15972	Specify the red, green, blue and alpha fade. Default values are @code{15},
15973	@code{10}, @code{5} and @code{5}.
15974	Allowed range is @code{[0, 255]}.
15975
15976	@item zoom
15977	Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
15978
15979	@item draw
15980	Set the vectorscope drawing mode.
15981
15982	Available values are:
15983	@table @samp
15984	@item dot
15985	Draw dot for each sample.
15986
15987	@item line
15988	Draw line between previous and current sample.
15989	@end table
15990
15991	Default value is @samp{dot}.
15992
15993	@item scale
15994	Specify amplitude scale of audio samples.
15995
15996	Available values are:
15997	@table @samp
15998	@item lin
15999	Linear.
16000
16001	@item sqrt
16002	Square root.
16003
16004	@item cbrt
16005	Cubic root.
16006
16007	@item log
16008	Logarithmic.
16009	@end table
16010
16011	@end table
16012
16013	@subsection Examples
16014
16015	@itemize
16016	@item
16017	Complete example using @command{ffplay}:
16018	@example
16019	ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
16020	[a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
16021	@end example
16022	@end itemize
16023
16024	@section bench, abench
16025
16026	Benchmark part of a filtergraph.
16027
16028	The filter accepts the following options:
16029
16030	@table @option
16031	@item action
16032	Start or stop a timer.
16033
16034	Available values are:
16035	@table @samp
16036	@item start
16037	Get the current time, set it as frame metadata (using the key
16038	@code{lavfi.bench.start_time}), and forward the frame to the next filter.
16039
16040	@item stop
16041	Get the current time and fetch the @code{lavfi.bench.start_time} metadata from
16042	the input frame metadata to get the time difference. Time difference, average,
16043	maximum and minimum time (respectively @code{t}, @code{avg}, @code{max} and
16044	@code{min}) are then printed. The timestamps are expressed in seconds.
16045	@end table
16046	@end table
16047
16048	@subsection Examples
16049
16050	@itemize
16051	@item
16052	Benchmark @ref{selectivecolor} filter:
16053	@example
16054	bench=start,selectivecolor=reds=-.2 .12 -.49,bench=stop
16055	@end example
16056	@end itemize
16057
16058	@section concat
16059
16060	Concatenate audio and video streams, joining them together one after the
16061	other.
16062
16063	The filter works on segments of synchronized video and audio streams. All
16064	segments must have the same number of streams of each type, and that will
16065	also be the number of streams at output.
16066
16067	The filter accepts the following options:
16068
16069	@table @option
16070
16071	@item n
16072	Set the number of segments. Default is 2.
16073
16074	@item v
16075	Set the number of output video streams, that is also the number of video
16076	streams in each segment. Default is 1.
16077
16078	@item a
16079	Set the number of output audio streams, that is also the number of audio
16080	streams in each segment. Default is 0.
16081
16082	@item unsafe
16083	Activate unsafe mode: do not fail if segments have a different format.
16084
16085	@end table
16086
16087	The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
16088	@var{a} audio outputs.
16089
16090	There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
16091	segment, in the same order as the outputs, then the inputs for the second
16092	segment, etc.
16093
16094	Related streams do not always have exactly the same duration, for various
16095	reasons including codec frame size or sloppy authoring. For that reason,
16096	related synchronized streams (e.g. a video and its audio track) should be
16097	concatenated at once. The concat filter will use the duration of the longest
16098	stream in each segment (except the last one), and if necessary pad shorter
16099	audio streams with silence.
16100
16101	For this filter to work correctly, all segments must start at timestamp 0.
16102
16103	All corresponding streams must have the same parameters in all segments; the
16104	filtering system will automatically select a common pixel format for video
16105	streams, and a common sample format, sample rate and channel layout for
16106	audio streams, but other settings, such as resolution, must be converted
16107	explicitly by the user.
16108
16109	Different frame rates are acceptable but will result in variable frame rate
16110	at output; be sure to configure the output file to handle it.
16111
16112	@subsection Examples
16113
16114	@itemize
16115	@item
16116	Concatenate an opening, an episode and an ending, all in bilingual version
16117	(video in stream 0, audio in streams 1 and 2):
16118	@example
16119	ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
16120	'[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
16121	concat=n=3:v=1:a=2 [v] [a1] [a2]' \
16122	-map '[v]' -map '[a1]' -map '[a2]' output.mkv
16123	@end example
16124
16125	@item
16126	Concatenate two parts, handling audio and video separately, using the
16127	(a)movie sources, and adjusting the resolution:
16128	@example
16129	movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
16130	movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
16131	[v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
16132	@end example
16133	Note that a desync will happen at the stitch if the audio and video streams
16134	do not have exactly the same duration in the first file.
16135
16136	@end itemize
16137
16138	@section drawgraph, adrawgraph
16139
16140	Draw a graph using input video or audio metadata.
16141
16142	It accepts the following parameters:
16143
16144	@table @option
16145	@item m1
16146	Set 1st frame metadata key from which metadata values will be used to draw a graph.
16147
16148	@item fg1
16149	Set 1st foreground color expression.
16150
16151	@item m2
16152	Set 2nd frame metadata key from which metadata values will be used to draw a graph.
16153
16154	@item fg2
16155	Set 2nd foreground color expression.
16156
16157	@item m3
16158	Set 3rd frame metadata key from which metadata values will be used to draw a graph.
16159
16160	@item fg3
16161	Set 3rd foreground color expression.
16162
16163	@item m4
16164	Set 4th frame metadata key from which metadata values will be used to draw a graph.
16165
16166	@item fg4
16167	Set 4th foreground color expression.
16168
16169	@item min
16170	Set minimal value of metadata value.
16171
16172	@item max
16173	Set maximal value of metadata value.
16174
16175	@item bg
16176	Set graph background color. Default is white.
16177
16178	@item mode
16179	Set graph mode.
16180
16181	Available values for mode is:
16182	@table @samp
16183	@item bar
16184	@item dot
16185	@item line
16186	@end table
16187
16188	Default is @code{line}.
16189
16190	@item slide
16191	Set slide mode.
16192
16193	Available values for slide is:
16194	@table @samp
16195	@item frame
16196	Draw new frame when right border is reached.
16197
16198	@item replace
16199	Replace old columns with new ones.
16200
16201	@item scroll
16202	Scroll from right to left.
16203
16204	@item rscroll
16205	Scroll from left to right.
16206
16207	@item picture
16208	Draw single picture.
16209	@end table
16210
16211	Default is @code{frame}.
16212
16213	@item size
16214	Set size of graph video. For the syntax of this option, check the
16215	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16216	The default value is @code{900x256}.
16217
16218	The foreground color expressions can use the following variables:
16219	@table @option
16220	@item MIN
16221	Minimal value of metadata value.
16222
16223	@item MAX
16224	Maximal value of metadata value.
16225
16226	@item VAL
16227	Current metadata key value.
16228	@end table
16229
16230	The color is defined as 0xAABBGGRR.
16231	@end table
16232
16233	Example using metadata from @ref{signalstats} filter:
16234	@example
16235	signalstats,drawgraph=lavfi.signalstats.YAVG:min=0:max=255
16236	@end example
16237
16238	Example using metadata from @ref{ebur128} filter:
16239	@example
16240	ebur128=metadata=1,adrawgraph=lavfi.r128.M:min=-120:max=5
16241	@end example
16242
16243	@anchor{ebur128}
16244	@section ebur128
16245
16246	EBU R128 scanner filter. This filter takes an audio stream as input and outputs
16247	it unchanged. By default, it logs a message at a frequency of 10Hz with the
16248	Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
16249	Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
16250
16251	The filter also has a video output (see the @var{video} option) with a real
16252	time graph to observe the loudness evolution. The graphic contains the logged
16253	message mentioned above, so it is not printed anymore when this option is set,
16254	unless the verbose logging is set. The main graphing area contains the
16255	short-term loudness (3 seconds of analysis), and the gauge on the right is for
16256	the momentary loudness (400 milliseconds).
16257
16258	More information about the Loudness Recommendation EBU R128 on
16259	@url{http://tech.ebu.ch/loudness}.
16260
16261	The filter accepts the following options:
16262
16263	@table @option
16264
16265	@item video
16266	Activate the video output. The audio stream is passed unchanged whether this
16267	option is set or no. The video stream will be the first output stream if
16268	activated. Default is @code{0}.
16269
16270	@item size
16271	Set the video size. This option is for video only. For the syntax of this
16272	option, check the
16273	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16274	Default and minimum resolution is @code{640x480}.
16275
16276	@item meter
16277	Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
16278	@code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
16279	other integer value between this range is allowed.
16280
16281	@item metadata
16282	Set metadata injection. If set to @code{1}, the audio input will be segmented
16283	into 100ms output frames, each of them containing various loudness information
16284	in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
16285
16286	Default is @code{0}.
16287
16288	@item framelog
16289	Force the frame logging level.
16290
16291	Available values are:
16292	@table @samp
16293	@item info
16294	information logging level
16295	@item verbose
16296	verbose logging level
16297	@end table
16298
16299	By default, the logging level is set to @var{info}. If the @option{video} or
16300	the @option{metadata} options are set, it switches to @var{verbose}.
16301
16302	@item peak
16303	Set peak mode(s).
16304
16305	Available modes can be cumulated (the option is a @code{flag} type). Possible
16306	values are:
16307	@table @samp
16308	@item none
16309	Disable any peak mode (default).
16310	@item sample
16311	Enable sample-peak mode.
16312
16313	Simple peak mode looking for the higher sample value. It logs a message
16314	for sample-peak (identified by @code{SPK}).
16315	@item true
16316	Enable true-peak mode.
16317
16318	If enabled, the peak lookup is done on an over-sampled version of the input
16319	stream for better peak accuracy. It logs a message for true-peak.
16320	(identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
16321	This mode requires a build with @code{libswresample}.
16322	@end table
16323
16324	@item dualmono
16325	Treat mono input files as "dual mono". If a mono file is intended for playback
16326	on a stereo system, its EBU R128 measurement will be perceptually incorrect.
16327	If set to @code{true}, this option will compensate for this effect.
16328	Multi-channel input files are not affected by this option.
16329
16330	@item panlaw
16331	Set a specific pan law to be used for the measurement of dual mono files.
16332	This parameter is optional, and has a default value of -3.01dB.
16333	@end table
16334
16335	@subsection Examples
16336
16337	@itemize
16338	@item
16339	Real-time graph using @command{ffplay}, with a EBU scale meter +18:
16340	@example
16341	ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
16342	@end example
16343
16344	@item
16345	Run an analysis with @command{ffmpeg}:
16346	@example
16347	ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
16348	@end example
16349	@end itemize
16350
16351	@section interleave, ainterleave
16352
16353	Temporally interleave frames from several inputs.
16354
16355	@code{interleave} works with video inputs, @code{ainterleave} with audio.
16356
16357	These filters read frames from several inputs and send the oldest
16358	queued frame to the output.
16359
16360	Input streams must have well defined, monotonically increasing frame
16361	timestamp values.
16362
16363	In order to submit one frame to output, these filters need to enqueue
16364	at least one frame for each input, so they cannot work in case one
16365	input is not yet terminated and will not receive incoming frames.
16366
16367	For example consider the case when one input is a @code{select} filter
16368	which always drops input frames. The @code{interleave} filter will keep
16369	reading from that input, but it will never be able to send new frames
16370	to output until the input sends an end-of-stream signal.
16371
16372	Also, depending on inputs synchronization, the filters will drop
16373	frames in case one input receives more frames than the other ones, and
16374	the queue is already filled.
16375
16376	These filters accept the following options:
16377
16378	@table @option
16379	@item nb_inputs, n
16380	Set the number of different inputs, it is 2 by default.
16381	@end table
16382
16383	@subsection Examples
16384
16385	@itemize
16386	@item
16387	Interleave frames belonging to different streams using @command{ffmpeg}:
16388	@example
16389	ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
16390	@end example
16391
16392	@item
16393	Add flickering blur effect:
16394	@example
16395	select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
16396	@end example
16397	@end itemize
16398
16399	@section metadata, ametadata
16400
16401	Manipulate frame metadata.
16402
16403	This filter accepts the following options:
16404
16405	@table @option
16406	@item mode
16407	Set mode of operation of the filter.
16408
16409	Can be one of the following:
16410
16411	@table @samp
16412	@item select
16413	If both @code{value} and @code{key} is set, select frames
16414	which have such metadata. If only @code{key} is set, select
16415	every frame that has such key in metadata.
16416
16417	@item add
16418	Add new metadata @code{key} and @code{value}. If key is already available
16419	do nothing.
16420
16421	@item modify
16422	Modify value of already present key.
16423
16424	@item delete
16425	If @code{value} is set, delete only keys that have such value.
16426	Otherwise, delete key. If @code{key} is not set, delete all metadata values in
16427	the frame.
16428
16429	@item print
16430	Print key and its value if metadata was found. If @code{key} is not set print all
16431	metadata values available in frame.
16432	@end table
16433
16434	@item key
16435	Set key used with all modes. Must be set for all modes except @code{print} and @code{delete}.
16436
16437	@item value
16438	Set metadata value which will be used. This option is mandatory for
16439	@code{modify} and @code{add} mode.
16440
16441	@item function
16442	Which function to use when comparing metadata value and @code{value}.
16443
16444	Can be one of following:
16445
16446	@table @samp
16447	@item same_str
16448	Values are interpreted as strings, returns true if metadata value is same as @code{value}.
16449
16450	@item starts_with
16451	Values are interpreted as strings, returns true if metadata value starts with
16452	the @code{value} option string.
16453
16454	@item less
16455	Values are interpreted as floats, returns true if metadata value is less than @code{value}.
16456
16457	@item equal
16458	Values are interpreted as floats, returns true if @code{value} is equal with metadata value.
16459
16460	@item greater
16461	Values are interpreted as floats, returns true if metadata value is greater than @code{value}.
16462
16463	@item expr
16464	Values are interpreted as floats, returns true if expression from option @code{expr}
16465	evaluates to true.
16466	@end table
16467
16468	@item expr
16469	Set expression which is used when @code{function} is set to @code{expr}.
16470	The expression is evaluated through the eval API and can contain the following
16471	constants:
16472
16473	@table @option
16474	@item VALUE1
16475	Float representation of @code{value} from metadata key.
16476
16477	@item VALUE2
16478	Float representation of @code{value} as supplied by user in @code{value} option.
16479	@end table
16480
16481	@item file
16482	If specified in @code{print} mode, output is written to the named file. Instead of
16483	plain filename any writable url can be specified. Filename ``-'' is a shorthand
16484	for standard output. If @code{file} option is not set, output is written to the log
16485	with AV_LOG_INFO loglevel.
16486
16487	@end table
16488
16489	@subsection Examples
16490
16491	@itemize
16492	@item
16493	Print all metadata values for frames with key @code{lavfi.singnalstats.YDIF} with values
16494	between 0 and 1.
16495	@example
16496	signalstats,metadata=print:key=lavfi.signalstats.YDIF:value=0:function=expr:expr='between(VALUE1,0,1)'
16497	@end example
16498	@item
16499	Print silencedetect output to file @file{metadata.txt}.
16500	@example
16501	silencedetect,ametadata=mode=print:file=metadata.txt
16502	@end example
16503	@item
16504	Direct all metadata to a pipe with file descriptor 4.
16505	@example
16506	metadata=mode=print:file='pipe\:4'
16507	@end example
16508	@end itemize
16509
16510	@section perms, aperms
16511
16512	Set read/write permissions for the output frames.
16513
16514	These filters are mainly aimed at developers to test direct path in the
16515	following filter in the filtergraph.
16516
16517	The filters accept the following options:
16518
16519	@table @option
16520	@item mode
16521	Select the permissions mode.
16522
16523	It accepts the following values:
16524	@table @samp
16525	@item none
16526	Do nothing. This is the default.
16527	@item ro
16528	Set all the output frames read-only.
16529	@item rw
16530	Set all the output frames directly writable.
16531	@item toggle
16532	Make the frame read-only if writable, and writable if read-only.
16533	@item random
16534	Set each output frame read-only or writable randomly.
16535	@end table
16536
16537	@item seed
16538	Set the seed for the @var{random} mode, must be an integer included between
16539	@code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
16540	@code{-1}, the filter will try to use a good random seed on a best effort
16541	basis.
16542	@end table
16543
16544	Note: in case of auto-inserted filter between the permission filter and the
16545	following one, the permission might not be received as expected in that
16546	following filter. Inserting a @ref{format} or @ref{aformat} filter before the
16547	perms/aperms filter can avoid this problem.
16548
16549	@section realtime, arealtime
16550
16551	Slow down filtering to match real time approximatively.
16552
16553	These filters will pause the filtering for a variable amount of time to
16554	match the output rate with the input timestamps.
16555	They are similar to the @option{re} option to @code{ffmpeg}.
16556
16557	They accept the following options:
16558
16559	@table @option
16560	@item limit
16561	Time limit for the pauses. Any pause longer than that will be considered
16562	a timestamp discontinuity and reset the timer. Default is 2 seconds.
16563	@end table
16564
16565	@anchor{select}
16566	@section select, aselect
16567
16568	Select frames to pass in output.
16569
16570	This filter accepts the following options:
16571
16572	@table @option
16573
16574	@item expr, e
16575	Set expression, which is evaluated for each input frame.
16576
16577	If the expression is evaluated to zero, the frame is discarded.
16578
16579	If the evaluation result is negative or NaN, the frame is sent to the
16580	first output; otherwise it is sent to the output with index
16581	@code{ceil(val)-1}, assuming that the input index starts from 0.
16582
16583	For example a value of @code{1.2} corresponds to the output with index
16584	@code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
16585
16586	@item outputs, n
16587	Set the number of outputs. The output to which to send the selected
16588	frame is based on the result of the evaluation. Default value is 1.
16589	@end table
16590
16591	The expression can contain the following constants:
16592
16593	@table @option
16594	@item n
16595	The (sequential) number of the filtered frame, starting from 0.
16596
16597	@item selected_n
16598	The (sequential) number of the selected frame, starting from 0.
16599
16600	@item prev_selected_n
16601	The sequential number of the last selected frame. It's NAN if undefined.
16602
16603	@item TB
16604	The timebase of the input timestamps.
16605
16606	@item pts
16607	The PTS (Presentation TimeStamp) of the filtered video frame,
16608	expressed in @var{TB} units. It's NAN if undefined.
16609
16610	@item t
16611	The PTS of the filtered video frame,
16612	expressed in seconds. It's NAN if undefined.
16613
16614	@item prev_pts
16615	The PTS of the previously filtered video frame. It's NAN if undefined.
16616
16617	@item prev_selected_pts
16618	The PTS of the last previously filtered video frame. It's NAN if undefined.
16619
16620	@item prev_selected_t
16621	The PTS of the last previously selected video frame. It's NAN if undefined.
16622
16623	@item start_pts
16624	The PTS of the first video frame in the video. It's NAN if undefined.
16625
16626	@item start_t
16627	The time of the first video frame in the video. It's NAN if undefined.
16628
16629	@item pict_type @emph{(video only)}
16630	The type of the filtered frame. It can assume one of the following
16631	values:
16632	@table @option
16633	@item I
16634	@item P
16635	@item B
16636	@item S
16637	@item SI
16638	@item SP
16639	@item BI
16640	@end table
16641
16642	@item interlace_type @emph{(video only)}
16643	The frame interlace type. It can assume one of the following values:
16644	@table @option
16645	@item PROGRESSIVE
16646	The frame is progressive (not interlaced).
16647	@item TOPFIRST
16648	The frame is top-field-first.
16649	@item BOTTOMFIRST
16650	The frame is bottom-field-first.
16651	@end table
16652
16653	@item consumed_sample_n @emph{(audio only)}
16654	the number of selected samples before the current frame
16655
16656	@item samples_n @emph{(audio only)}
16657	the number of samples in the current frame
16658
16659	@item sample_rate @emph{(audio only)}
16660	the input sample rate
16661
16662	@item key
16663	This is 1 if the filtered frame is a key-frame, 0 otherwise.
16664
16665	@item pos
16666	the position in the file of the filtered frame, -1 if the information
16667	is not available (e.g. for synthetic video)
16668
16669	@item scene @emph{(video only)}
16670	value between 0 and 1 to indicate a new scene; a low value reflects a low
16671	probability for the current frame to introduce a new scene, while a higher
16672	value means the current frame is more likely to be one (see the example below)
16673
16674	@item concatdec_select
16675	The concat demuxer can select only part of a concat input file by setting an
16676	inpoint and an outpoint, but the output packets may not be entirely contained
16677	in the selected interval. By using this variable, it is possible to skip frames
16678	generated by the concat demuxer which are not exactly contained in the selected
16679	interval.
16680
16681	This works by comparing the frame pts against the @var{lavf.concat.start_time}
16682	and the @var{lavf.concat.duration} packet metadata values which are also
16683	present in the decoded frames.
16684
16685	The @var{concatdec_select} variable is -1 if the frame pts is at least
16686	start_time and either the duration metadata is missing or the frame pts is less
16687	than start_time + duration, 0 otherwise, and NaN if the start_time metadata is
16688	missing.
16689
16690	That basically means that an input frame is selected if its pts is within the
16691	interval set by the concat demuxer.
16692
16693	@end table
16694
16695	The default value of the select expression is "1".
16696
16697	@subsection Examples
16698
16699	@itemize
16700	@item
16701	Select all frames in input:
16702	@example
16703	select
16704	@end example
16705
16706	The example above is the same as:
16707	@example
16708	select=1
16709	@end example
16710
16711	@item
16712	Skip all frames:
16713	@example
16714	select=0
16715	@end example
16716
16717	@item
16718	Select only I-frames:
16719	@example
16720	select='eq(pict_type\,I)'
16721	@end example
16722
16723	@item
16724	Select one frame every 100:
16725	@example
16726	select='not(mod(n\,100))'
16727	@end example
16728
16729	@item
16730	Select only frames contained in the 10-20 time interval:
16731	@example
16732	select=between(t\,10\,20)
16733	@end example
16734
16735	@item
16736	Select only I-frames contained in the 10-20 time interval:
16737	@example
16738	select=between(t\,10\,20)*eq(pict_type\,I)
16739	@end example
16740
16741	@item
16742	Select frames with a minimum distance of 10 seconds:
16743	@example
16744	select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
16745	@end example
16746
16747	@item
16748	Use aselect to select only audio frames with samples number > 100:
16749	@example
16750	aselect='gt(samples_n\,100)'
16751	@end example
16752
16753	@item
16754	Create a mosaic of the first scenes:
16755	@example
16756	ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
16757	@end example
16758
16759	Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
16760	choice.
16761
16762	@item
16763	Send even and odd frames to separate outputs, and compose them:
16764	@example
16765	select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
16766	@end example
16767
16768	@item
16769	Select useful frames from an ffconcat file which is using inpoints and
16770	outpoints but where the source files are not intra frame only.
16771	@example
16772	ffmpeg -copyts -vsync 0 -segment_time_metadata 1 -i input.ffconcat -vf select=concatdec_select -af aselect=concatdec_select output.avi
16773	@end example
16774	@end itemize
16775
16776	@section sendcmd, asendcmd
16777
16778	Send commands to filters in the filtergraph.
16779
16780	These filters read commands to be sent to other filters in the
16781	filtergraph.
16782
16783	@code{sendcmd} must be inserted between two video filters,
16784	@code{asendcmd} must be inserted between two audio filters, but apart
16785	from that they act the same way.
16786
16787	The specification of commands can be provided in the filter arguments
16788	with the @var{commands} option, or in a file specified by the
16789	@var{filename} option.
16790
16791	These filters accept the following options:
16792	@table @option
16793	@item commands, c
16794	Set the commands to be read and sent to the other filters.
16795	@item filename, f
16796	Set the filename of the commands to be read and sent to the other
16797	filters.
16798	@end table
16799
16800	@subsection Commands syntax
16801
16802	A commands description consists of a sequence of interval
16803	specifications, comprising a list of commands to be executed when a
16804	particular event related to that interval occurs. The occurring event
16805	is typically the current frame time entering or leaving a given time
16806	interval.
16807
16808	An interval is specified by the following syntax:
16809	@example
16810	@var{START}[-@var{END}] @var{COMMANDS};
16811	@end example
16812
16813	The time interval is specified by the @var{START} and @var{END} times.
16814	@var{END} is optional and defaults to the maximum time.
16815
16816	The current frame time is considered within the specified interval if
16817	it is included in the interval [@var{START}, @var{END}), that is when
16818	the time is greater or equal to @var{START} and is lesser than
16819	@var{END}.
16820
16821	@var{COMMANDS} consists of a sequence of one or more command
16822	specifications, separated by ",", relating to that interval. The
16823	syntax of a command specification is given by:
16824	@example
16825	[@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
16826	@end example
16827
16828	@var{FLAGS} is optional and specifies the type of events relating to
16829	the time interval which enable sending the specified command, and must
16830	be a non-null sequence of identifier flags separated by "+" or "|" and
16831	enclosed between "[" and "]".
16832
16833	The following flags are recognized:
16834	@table @option
16835	@item enter
16836	The command is sent when the current frame timestamp enters the
16837	specified interval. In other words, the command is sent when the
16838	previous frame timestamp was not in the given interval, and the
16839	current is.
16840
16841	@item leave
16842	The command is sent when the current frame timestamp leaves the
16843	specified interval. In other words, the command is sent when the
16844	previous frame timestamp was in the given interval, and the
16845	current is not.
16846	@end table
16847
16848	If @var{FLAGS} is not specified, a default value of @code{[enter]} is
16849	assumed.
16850
16851	@var{TARGET} specifies the target of the command, usually the name of
16852	the filter class or a specific filter instance name.
16853
16854	@var{COMMAND} specifies the name of the command for the target filter.
16855
16856	@var{ARG} is optional and specifies the optional list of argument for
16857	the given @var{COMMAND}.
16858
16859	Between one interval specification and another, whitespaces, or
16860	sequences of characters starting with @code{#} until the end of line,
16861	are ignored and can be used to annotate comments.
16862
16863	A simplified BNF description of the commands specification syntax
16864	follows:
16865	@example
16866	@var{COMMAND_FLAG} ::= "enter" | "leave"
16867	@var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
16868	@var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
16869	@var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
16870	@var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
16871	@var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
16872	@end example
16873
16874	@subsection Examples
16875
16876	@itemize
16877	@item
16878	Specify audio tempo change at second 4:
16879	@example
16880	asendcmd=c='4.0 atempo tempo 1.5',atempo
16881	@end example
16882
16883	@item
16884	Specify a list of drawtext and hue commands in a file.
16885	@example
16886	# show text in the interval 5-10
16887	5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
16888	[leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
16889
16890	# desaturate the image in the interval 15-20
16891	15.0-20.0 [enter] hue s 0,
16892	[enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
16893	[leave] hue s 1,
16894	[leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
16895
16896	# apply an exponential saturation fade-out effect, starting from time 25
16897	25 [enter] hue s exp(25-t)
16898	@end example
16899
16900	A filtergraph allowing to read and process the above command list
16901	stored in a file @file{test.cmd}, can be specified with:
16902	@example
16903	sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
16904	@end example
16905	@end itemize
16906
16907	@anchor{setpts}
16908	@section setpts, asetpts
16909
16910	Change the PTS (presentation timestamp) of the input frames.
16911
16912	@code{setpts} works on video frames, @code{asetpts} on audio frames.
16913
16914	This filter accepts the following options:
16915
16916	@table @option
16917
16918	@item expr
16919	The expression which is evaluated for each frame to construct its timestamp.
16920
16921	@end table
16922
16923	The expression is evaluated through the eval API and can contain the following
16924	constants:
16925
16926	@table @option
16927	@item FRAME_RATE
16928	frame rate, only defined for constant frame-rate video
16929
16930	@item PTS
16931	The presentation timestamp in input
16932
16933	@item N
16934	The count of the input frame for video or the number of consumed samples,
16935	not including the current frame for audio, starting from 0.
16936
16937	@item NB_CONSUMED_SAMPLES
16938	The number of consumed samples, not including the current frame (only
16939	audio)
16940
16941	@item NB_SAMPLES, S
16942	The number of samples in the current frame (only audio)
16943
16944	@item SAMPLE_RATE, SR
16945	The audio sample rate.
16946
16947	@item STARTPTS
16948	The PTS of the first frame.
16949
16950	@item STARTT
16951	the time in seconds of the first frame
16952
16953	@item INTERLACED
16954	State whether the current frame is interlaced.
16955
16956	@item T
16957	the time in seconds of the current frame
16958
16959	@item POS
16960	original position in the file of the frame, or undefined if undefined
16961	for the current frame
16962
16963	@item PREV_INPTS
16964	The previous input PTS.
16965
16966	@item PREV_INT
16967	previous input time in seconds
16968
16969	@item PREV_OUTPTS
16970	The previous output PTS.
16971
16972	@item PREV_OUTT
16973	previous output time in seconds
16974
16975	@item RTCTIME
16976	The wallclock (RTC) time in microseconds. This is deprecated, use time(0)
16977	instead.
16978
16979	@item RTCSTART
16980	The wallclock (RTC) time at the start of the movie in microseconds.
16981
16982	@item TB
16983	The timebase of the input timestamps.
16984
16985	@end table
16986
16987	@subsection Examples
16988
16989	@itemize
16990	@item
16991	Start counting PTS from zero
16992	@example
16993	setpts=PTS-STARTPTS
16994	@end example
16995
16996	@item
16997	Apply fast motion effect:
16998	@example
16999	setpts=0.5*PTS
17000	@end example
17001
17002	@item
17003	Apply slow motion effect:
17004	@example
17005	setpts=2.0*PTS
17006	@end example
17007
17008	@item
17009	Set fixed rate of 25 frames per second:
17010	@example
17011	setpts=N/(25*TB)
17012	@end example
17013
17014	@item
17015	Set fixed rate 25 fps with some jitter:
17016	@example
17017	setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
17018	@end example
17019
17020	@item
17021	Apply an offset of 10 seconds to the input PTS:
17022	@example
17023	setpts=PTS+10/TB
17024	@end example
17025
17026	@item
17027	Generate timestamps from a "live source" and rebase onto the current timebase:
17028	@example
17029	setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
17030	@end example
17031
17032	@item
17033	Generate timestamps by counting samples:
17034	@example
17035	asetpts=N/SR/TB
17036	@end example
17037
17038	@end itemize
17039
17040	@section settb, asettb
17041
17042	Set the timebase to use for the output frames timestamps.
17043	It is mainly useful for testing timebase configuration.
17044
17045	It accepts the following parameters:
17046
17047	@table @option
17048
17049	@item expr, tb
17050	The expression which is evaluated into the output timebase.
17051
17052	@end table
17053
17054	The value for @option{tb} is an arithmetic expression representing a
17055	rational. The expression can contain the constants "AVTB" (the default
17056	timebase), "intb" (the input timebase) and "sr" (the sample rate,
17057	audio only). Default value is "intb".
17058
17059	@subsection Examples
17060
17061	@itemize
17062	@item
17063	Set the timebase to 1/25:
17064	@example
17065	settb=expr=1/25
17066	@end example
17067
17068	@item
17069	Set the timebase to 1/10:
17070	@example
17071	settb=expr=0.1
17072	@end example
17073
17074	@item
17075	Set the timebase to 1001/1000:
17076	@example
17077	settb=1+0.001
17078	@end example
17079
17080	@item
17081	Set the timebase to 2*intb:
17082	@example
17083	settb=2*intb
17084	@end example
17085
17086	@item
17087	Set the default timebase value:
17088	@example
17089	settb=AVTB
17090	@end example
17091	@end itemize
17092
17093	@section showcqt
17094	Convert input audio to a video output representing frequency spectrum
17095	logarithmically using Brown-Puckette constant Q transform algorithm with
17096	direct frequency domain coefficient calculation (but the transform itself
17097	is not really constant Q, instead the Q factor is actually variable/clamped),
17098	with musical tone scale, from E0 to D#10.
17099
17100	The filter accepts the following options:
17101
17102	@table @option
17103	@item size, s
17104	Specify the video size for the output. It must be even. For the syntax of this option,
17105	check the @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17106	Default value is @code{1920x1080}.
17107
17108	@item fps, rate, r
17109	Set the output frame rate. Default value is @code{25}.
17110
17111	@item bar_h
17112	Set the bargraph height. It must be even. Default value is @code{-1} which
17113	computes the bargraph height automatically.
17114
17115	@item axis_h
17116	Set the axis height. It must be even. Default value is @code{-1} which computes
17117	the axis height automatically.
17118
17119	@item sono_h
17120	Set the sonogram height. It must be even. Default value is @code{-1} which
17121	computes the sonogram height automatically.
17122
17123	@item fullhd
17124	Set the fullhd resolution. This option is deprecated, use @var{size}, @var{s}
17125	instead. Default value is @code{1}.
17126
17127	@item sono_v, volume
17128	Specify the sonogram volume expression. It can contain variables:
17129	@table @option
17130	@item bar_v
17131	the @var{bar_v} evaluated expression
17132	@item frequency, freq, f
17133	the frequency where it is evaluated
17134	@item timeclamp, tc
17135	the value of @var{timeclamp} option
17136	@end table
17137	and functions:
17138	@table @option
17139	@item a_weighting(f)
17140	A-weighting of equal loudness
17141	@item b_weighting(f)
17142	B-weighting of equal loudness
17143	@item c_weighting(f)
17144	C-weighting of equal loudness.
17145	@end table
17146	Default value is @code{16}.
17147
17148	@item bar_v, volume2
17149	Specify the bargraph volume expression. It can contain variables:
17150	@table @option
17151	@item sono_v
17152	the @var{sono_v} evaluated expression
17153	@item frequency, freq, f
17154	the frequency where it is evaluated
17155	@item timeclamp, tc
17156	the value of @var{timeclamp} option
17157	@end table
17158	and functions:
17159	@table @option
17160	@item a_weighting(f)
17161	A-weighting of equal loudness
17162	@item b_weighting(f)
17163	B-weighting of equal loudness
17164	@item c_weighting(f)
17165	C-weighting of equal loudness.
17166	@end table
17167	Default value is @code{sono_v}.
17168
17169	@item sono_g, gamma
17170	Specify the sonogram gamma. Lower gamma makes the spectrum more contrast,
17171	higher gamma makes the spectrum having more range. Default value is @code{3}.
17172	Acceptable range is @code{[1, 7]}.
17173
17174	@item bar_g, gamma2
17175	Specify the bargraph gamma. Default value is @code{1}. Acceptable range is
17176	@code{[1, 7]}.
17177
17178	@item bar_t
17179	Specify the bargraph transparency level. Lower value makes the bargraph sharper.
17180	Default value is @code{1}. Acceptable range is @code{[0, 1]}.
17181
17182	@item timeclamp, tc
17183	Specify the transform timeclamp. At low frequency, there is trade-off between
17184	accuracy in time domain and frequency domain. If timeclamp is lower,
17185	event in time domain is represented more accurately (such as fast bass drum),
17186	otherwise event in frequency domain is represented more accurately
17187	(such as bass guitar). Acceptable range is @code{[0.002, 1]}. Default value is @code{0.17}.
17188
17189	@item basefreq
17190	Specify the transform base frequency. Default value is @code{20.01523126408007475},
17191	which is frequency 50 cents below E0. Acceptable range is @code{[10, 100000]}.
17192
17193	@item endfreq
17194	Specify the transform end frequency. Default value is @code{20495.59681441799654},
17195	which is frequency 50 cents above D#10. Acceptable range is @code{[10, 100000]}.
17196
17197	@item coeffclamp
17198	This option is deprecated and ignored.
17199
17200	@item tlength
17201	Specify the transform length in time domain. Use this option to control accuracy
17202	trade-off between time domain and frequency domain at every frequency sample.
17203	It can contain variables:
17204	@table @option
17205	@item frequency, freq, f
17206	the frequency where it is evaluated
17207	@item timeclamp, tc
17208	the value of @var{timeclamp} option.
17209	@end table
17210	Default value is @code{384*tc/(384+tc*f)}.
17211
17212	@item count
17213	Specify the transform count for every video frame. Default value is @code{6}.
17214	Acceptable range is @code{[1, 30]}.
17215
17216	@item fcount
17217	Specify the transform count for every single pixel. Default value is @code{0},
17218	which makes it computed automatically. Acceptable range is @code{[0, 10]}.
17219
17220	@item fontfile
17221	Specify font file for use with freetype to draw the axis. If not specified,
17222	use embedded font. Note that drawing with font file or embedded font is not
17223	implemented with custom @var{basefreq} and @var{endfreq}, use @var{axisfile}
17224	option instead.
17225
17226	@item font
17227	Specify fontconfig pattern. This has lower priority than @var{fontfile}.
17228	The : in the pattern may be replaced by | to avoid unnecessary escaping.
17229
17230	@item fontcolor
17231	Specify font color expression. This is arithmetic expression that should return
17232	integer value 0xRRGGBB. It can contain variables:
17233	@table @option
17234	@item frequency, freq, f
17235	the frequency where it is evaluated
17236	@item timeclamp, tc
17237	the value of @var{timeclamp} option
17238	@end table
17239	and functions:
17240	@table @option
17241	@item midi(f)
17242	midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
17243	@item r(x), g(x), b(x)
17244	red, green, and blue value of intensity x.
17245	@end table
17246	Default value is @code{st(0, (midi(f)-59.5)/12);
17247	st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
17248	r(1-ld(1)) + b(ld(1))}.
17249
17250	@item axisfile
17251	Specify image file to draw the axis. This option override @var{fontfile} and
17252	@var{fontcolor} option.
17253
17254	@item axis, text
17255	Enable/disable drawing text to the axis. If it is set to @code{0}, drawing to
17256	the axis is disabled, ignoring @var{fontfile} and @var{axisfile} option.
17257	Default value is @code{1}.
17258
17259	@item csp
17260	Set colorspace. The accepted values are:
17261	@table @samp
17262	@item unspecified
17263	Unspecified (default)
17264
17265	@item bt709
17266	BT.709
17267
17268	@item fcc
17269	FCC
17270
17271	@item bt470bg
17272	BT.470BG or BT.601-6 625
17273
17274	@item smpte170m
17275	SMPTE-170M or BT.601-6 525
17276
17277	@item smpte240m
17278	SMPTE-240M
17279
17280	@item bt2020ncl
17281	BT.2020 with non-constant luminance
17282
17283	@end table
17284
17285	@item cscheme
17286	Set spectrogram color scheme. This is list of floating point values with format
17287	@code{left_r|left_g|left_b|right_r|right_g|right_b}.
17288	The default is @code{1|0.5|0|0|0.5|1}.
17289
17290	@end table
17291
17292	@subsection Examples
17293
17294	@itemize
17295	@item
17296	Playing audio while showing the spectrum:
17297	@example
17298	ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
17299	@end example
17300
17301	@item
17302	Same as above, but with frame rate 30 fps:
17303	@example
17304	ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
17305	@end example
17306
17307	@item
17308	Playing at 1280x720:
17309	@example
17310	ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=s=1280x720:count=4 [out0]'
17311	@end example
17312
17313	@item
17314	Disable sonogram display:
17315	@example
17316	sono_h=0
17317	@end example
17318
17319	@item
17320	A1 and its harmonics: A1, A2, (near)E3, A3:
17321	@example
17322	ffplay -f lavfi 'aevalsrc=0.1*sin(2*PI*55*t)+0.1*sin(4*PI*55*t)+0.1*sin(6*PI*55*t)+0.1*sin(8*PI*55*t),
17323	asplit[a][out1]; [a] showcqt [out0]'
17324	@end example
17325
17326	@item
17327	Same as above, but with more accuracy in frequency domain:
17328	@example
17329	ffplay -f lavfi 'aevalsrc=0.1*sin(2*PI*55*t)+0.1*sin(4*PI*55*t)+0.1*sin(6*PI*55*t)+0.1*sin(8*PI*55*t),
17330	asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
17331	@end example
17332
17333	@item
17334	Custom volume:
17335	@example
17336	bar_v=10:sono_v=bar_v*a_weighting(f)
17337	@end example
17338
17339	@item
17340	Custom gamma, now spectrum is linear to the amplitude.
17341	@example
17342	bar_g=2:sono_g=2
17343	@end example
17344
17345	@item
17346	Custom tlength equation:
17347	@example
17348	tc=0.33:tlength='st(0,0.17); 384*tc / (384 / ld(0) + tc*f /(1-ld(0))) + 384*tc / (tc*f / ld(0) + 384 /(1-ld(0)))'
17349	@end example
17350
17351	@item
17352	Custom fontcolor and fontfile, C-note is colored green, others are colored blue:
17353	@example
17354	fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))':fontfile=myfont.ttf
17355	@end example
17356
17357	@item
17358	Custom font using fontconfig:
17359	@example
17360	font='Courier New,Monospace,mono|bold'
17361	@end example
17362
17363	@item
17364	Custom frequency range with custom axis using image file:
17365	@example
17366	axisfile=myaxis.png:basefreq=40:endfreq=10000
17367	@end example
17368	@end itemize
17369
17370	@section showfreqs
17371
17372	Convert input audio to video output representing the audio power spectrum.
17373	Audio amplitude is on Y-axis while frequency is on X-axis.
17374
17375	The filter accepts the following options:
17376
17377	@table @option
17378	@item size, s
17379	Specify size of video. For the syntax of this option, check the
17380	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17381	Default is @code{1024x512}.
17382
17383	@item mode
17384	Set display mode.
17385	This set how each frequency bin will be represented.
17386
17387	It accepts the following values:
17388	@table @samp
17389	@item line
17390	@item bar
17391	@item dot
17392	@end table
17393	Default is @code{bar}.
17394
17395	@item ascale
17396	Set amplitude scale.
17397
17398	It accepts the following values:
17399	@table @samp
17400	@item lin
17401	Linear scale.
17402
17403	@item sqrt
17404	Square root scale.
17405
17406	@item cbrt
17407	Cubic root scale.
17408
17409	@item log
17410	Logarithmic scale.
17411	@end table
17412	Default is @code{log}.
17413
17414	@item fscale
17415	Set frequency scale.
17416
17417	It accepts the following values:
17418	@table @samp
17419	@item lin
17420	Linear scale.
17421
17422	@item log
17423	Logarithmic scale.
17424
17425	@item rlog
17426	Reverse logarithmic scale.
17427	@end table
17428	Default is @code{lin}.
17429
17430	@item win_size
17431	Set window size.
17432
17433	It accepts the following values:
17434	@table @samp
17435	@item w16
17436	@item w32
17437	@item w64
17438	@item w128
17439	@item w256
17440	@item w512
17441	@item w1024
17442	@item w2048
17443	@item w4096
17444	@item w8192
17445	@item w16384
17446	@item w32768
17447	@item w65536
17448	@end table
17449	Default is @code{w2048}
17450
17451	@item win_func
17452	Set windowing function.
17453
17454	It accepts the following values:
17455	@table @samp
17456	@item rect
17457	@item bartlett
17458	@item hanning
17459	@item hamming
17460	@item blackman
17461	@item welch
17462	@item flattop
17463	@item bharris
17464	@item bnuttall
17465	@item bhann
17466	@item sine
17467	@item nuttall
17468	@item lanczos
17469	@item gauss
17470	@item tukey
17471	@item dolph
17472	@item cauchy
17473	@item parzen
17474	@item poisson
17475	@end table
17476	Default is @code{hanning}.
17477
17478	@item overlap
17479	Set window overlap. In range @code{[0, 1]}. Default is @code{1},
17480	which means optimal overlap for selected window function will be picked.
17481
17482	@item averaging
17483	Set time averaging. Setting this to 0 will display current maximal peaks.
17484	Default is @code{1}, which means time averaging is disabled.
17485
17486	@item colors
17487	Specify list of colors separated by space or by '|' which will be used to
17488	draw channel frequencies. Unrecognized or missing colors will be replaced
17489	by white color.
17490
17491	@item cmode
17492	Set channel display mode.
17493
17494	It accepts the following values:
17495	@table @samp
17496	@item combined
17497	@item separate
17498	@end table
17499	Default is @code{combined}.
17500
17501	@item minamp
17502	Set minimum amplitude used in @code{log} amplitude scaler.
17503
17504	@end table
17505
17506	@anchor{showspectrum}
17507	@section showspectrum
17508
17509	Convert input audio to a video output, representing the audio frequency
17510	spectrum.
17511
17512	The filter accepts the following options:
17513
17514	@table @option
17515	@item size, s
17516	Specify the video size for the output. For the syntax of this option, check the
17517	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17518	Default value is @code{640x512}.
17519
17520	@item slide
17521	Specify how the spectrum should slide along the window.
17522
17523	It accepts the following values:
17524	@table @samp
17525	@item replace
17526	the samples start again on the left when they reach the right
17527	@item scroll
17528	the samples scroll from right to left
17529	@item fullframe
17530	frames are only produced when the samples reach the right
17531	@item rscroll
17532	the samples scroll from left to right
17533	@end table
17534
17535	Default value is @code{replace}.
17536
17537	@item mode
17538	Specify display mode.
17539
17540	It accepts the following values:
17541	@table @samp
17542	@item combined
17543	all channels are displayed in the same row
17544	@item separate
17545	all channels are displayed in separate rows
17546	@end table
17547
17548	Default value is @samp{combined}.
17549
17550	@item color
17551	Specify display color mode.
17552
17553	It accepts the following values:
17554	@table @samp
17555	@item channel
17556	each channel is displayed in a separate color
17557	@item intensity
17558	each channel is displayed using the same color scheme
17559	@item rainbow
17560	each channel is displayed using the rainbow color scheme
17561	@item moreland
17562	each channel is displayed using the moreland color scheme
17563	@item nebulae
17564	each channel is displayed using the nebulae color scheme
17565	@item fire
17566	each channel is displayed using the fire color scheme
17567	@item fiery
17568	each channel is displayed using the fiery color scheme
17569	@item fruit
17570	each channel is displayed using the fruit color scheme
17571	@item cool
17572	each channel is displayed using the cool color scheme
17573	@end table
17574
17575	Default value is @samp{channel}.
17576
17577	@item scale
17578	Specify scale used for calculating intensity color values.
17579
17580	It accepts the following values:
17581	@table @samp
17582	@item lin
17583	linear
17584	@item sqrt
17585	square root, default
17586	@item cbrt
17587	cubic root
17588	@item log
17589	logarithmic
17590	@item 4thrt
17591	4th root
17592	@item 5thrt
17593	5th root
17594	@end table
17595
17596	Default value is @samp{sqrt}.
17597
17598	@item saturation
17599	Set saturation modifier for displayed colors. Negative values provide
17600	alternative color scheme. @code{0} is no saturation at all.
17601	Saturation must be in [-10.0, 10.0] range.
17602	Default value is @code{1}.
17603
17604	@item win_func
17605	Set window function.
17606
17607	It accepts the following values:
17608	@table @samp
17609	@item rect
17610	@item bartlett
17611	@item hann
17612	@item hanning
17613	@item hamming
17614	@item blackman
17615	@item welch
17616	@item flattop
17617	@item bharris
17618	@item bnuttall
17619	@item bhann
17620	@item sine
17621	@item nuttall
17622	@item lanczos
17623	@item gauss
17624	@item tukey
17625	@item dolph
17626	@item cauchy
17627	@item parzen
17628	@item poisson
17629	@end table
17630
17631	Default value is @code{hann}.
17632
17633	@item orientation
17634	Set orientation of time vs frequency axis. Can be @code{vertical} or
17635	@code{horizontal}. Default is @code{vertical}.
17636
17637	@item overlap
17638	Set ratio of overlap window. Default value is @code{0}.
17639	When value is @code{1} overlap is set to recommended size for specific
17640	window function currently used.
17641
17642	@item gain
17643	Set scale gain for calculating intensity color values.
17644	Default value is @code{1}.
17645
17646	@item data
17647	Set which data to display. Can be @code{magnitude}, default or @code{phase}.
17648
17649	@item rotation
17650	Set color rotation, must be in [-1.0, 1.0] range.
17651	Default value is @code{0}.
17652	@end table
17653
17654	The usage is very similar to the showwaves filter; see the examples in that
17655	section.
17656
17657	@subsection Examples
17658
17659	@itemize
17660	@item
17661	Large window with logarithmic color scaling:
17662	@example
17663	showspectrum=s=1280x480:scale=log
17664	@end example
17665
17666	@item
17667	Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
17668	@example
17669	ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
17670	[a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
17671	@end example
17672	@end itemize
17673
17674	@section showspectrumpic
17675
17676	Convert input audio to a single video frame, representing the audio frequency
17677	spectrum.
17678
17679	The filter accepts the following options:
17680
17681	@table @option
17682	@item size, s
17683	Specify the video size for the output. For the syntax of this option, check the
17684	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17685	Default value is @code{4096x2048}.
17686
17687	@item mode
17688	Specify display mode.
17689
17690	It accepts the following values:
17691	@table @samp
17692	@item combined
17693	all channels are displayed in the same row
17694	@item separate
17695	all channels are displayed in separate rows
17696	@end table
17697	Default value is @samp{combined}.
17698
17699	@item color
17700	Specify display color mode.
17701
17702	It accepts the following values:
17703	@table @samp
17704	@item channel
17705	each channel is displayed in a separate color
17706	@item intensity
17707	each channel is displayed using the same color scheme
17708	@item rainbow
17709	each channel is displayed using the rainbow color scheme
17710	@item moreland
17711	each channel is displayed using the moreland color scheme
17712	@item nebulae
17713	each channel is displayed using the nebulae color scheme
17714	@item fire
17715	each channel is displayed using the fire color scheme
17716	@item fiery
17717	each channel is displayed using the fiery color scheme
17718	@item fruit
17719	each channel is displayed using the fruit color scheme
17720	@item cool
17721	each channel is displayed using the cool color scheme
17722	@end table
17723	Default value is @samp{intensity}.
17724
17725	@item scale
17726	Specify scale used for calculating intensity color values.
17727
17728	It accepts the following values:
17729	@table @samp
17730	@item lin
17731	linear
17732	@item sqrt
17733	square root, default
17734	@item cbrt
17735	cubic root
17736	@item log
17737	logarithmic
17738	@item 4thrt
17739	4th root
17740	@item 5thrt
17741	5th root
17742	@end table
17743	Default value is @samp{log}.
17744
17745	@item saturation
17746	Set saturation modifier for displayed colors. Negative values provide
17747	alternative color scheme. @code{0} is no saturation at all.
17748	Saturation must be in [-10.0, 10.0] range.
17749	Default value is @code{1}.
17750
17751	@item win_func
17752	Set window function.
17753
17754	It accepts the following values:
17755	@table @samp
17756	@item rect
17757	@item bartlett
17758	@item hann
17759	@item hanning
17760	@item hamming
17761	@item blackman
17762	@item welch
17763	@item flattop
17764	@item bharris
17765	@item bnuttall
17766	@item bhann
17767	@item sine
17768	@item nuttall
17769	@item lanczos
17770	@item gauss
17771	@item tukey
17772	@item dolph
17773	@item cauchy
17774	@item parzen
17775	@item poisson
17776	@end table
17777	Default value is @code{hann}.
17778
17779	@item orientation
17780	Set orientation of time vs frequency axis. Can be @code{vertical} or
17781	@code{horizontal}. Default is @code{vertical}.
17782
17783	@item gain
17784	Set scale gain for calculating intensity color values.
17785	Default value is @code{1}.
17786
17787	@item legend
17788	Draw time and frequency axes and legends. Default is enabled.
17789
17790	@item rotation
17791	Set color rotation, must be in [-1.0, 1.0] range.
17792	Default value is @code{0}.
17793	@end table
17794
17795	@subsection Examples
17796
17797	@itemize
17798	@item
17799	Extract an audio spectrogram of a whole audio track
17800	in a 1024x1024 picture using @command{ffmpeg}:
17801	@example
17802	ffmpeg -i audio.flac -lavfi showspectrumpic=s=1024x1024 spectrogram.png
17803	@end example
17804	@end itemize
17805
17806	@section showvolume
17807
17808	Convert input audio volume to a video output.
17809
17810	The filter accepts the following options:
17811
17812	@table @option
17813	@item rate, r
17814	Set video rate.
17815
17816	@item b
17817	Set border width, allowed range is [0, 5]. Default is 1.
17818
17819	@item w
17820	Set channel width, allowed range is [80, 8192]. Default is 400.
17821
17822	@item h
17823	Set channel height, allowed range is [1, 900]. Default is 20.
17824
17825	@item f
17826	Set fade, allowed range is [0.001, 1]. Default is 0.95.
17827
17828	@item c
17829	Set volume color expression.
17830
17831	The expression can use the following variables:
17832
17833	@table @option
17834	@item VOLUME
17835	Current max volume of channel in dB.
17836
17837	@item PEAK
17838	Current peak.
17839
17840	@item CHANNEL
17841	Current channel number, starting from 0.
17842	@end table
17843
17844	@item t
17845	If set, displays channel names. Default is enabled.
17846
17847	@item v
17848	If set, displays volume values. Default is enabled.
17849
17850	@item o
17851	Set orientation, can be @code{horizontal} or @code{vertical},
17852	default is @code{horizontal}.
17853
17854	@item s
17855	Set step size, allowed range s [0, 5]. Default is 0, which means
17856	step is disabled.
17857	@end table
17858
17859	@section showwaves
17860
17861	Convert input audio to a video output, representing the samples waves.
17862
17863	The filter accepts the following options:
17864
17865	@table @option
17866	@item size, s
17867	Specify the video size for the output. For the syntax of this option, check the
17868	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17869	Default value is @code{600x240}.
17870
17871	@item mode
17872	Set display mode.
17873
17874	Available values are:
17875	@table @samp
17876	@item point
17877	Draw a point for each sample.
17878
17879	@item line
17880	Draw a vertical line for each sample.
17881
17882	@item p2p
17883	Draw a point for each sample and a line between them.
17884
17885	@item cline
17886	Draw a centered vertical line for each sample.
17887	@end table
17888
17889	Default value is @code{point}.
17890
17891	@item n
17892	Set the number of samples which are printed on the same column. A
17893	larger value will decrease the frame rate. Must be a positive
17894	integer. This option can be set only if the value for @var{rate}
17895	is not explicitly specified.
17896
17897	@item rate, r
17898	Set the (approximate) output frame rate. This is done by setting the
17899	option @var{n}. Default value is "25".
17900
17901	@item split_channels
17902	Set if channels should be drawn separately or overlap. Default value is 0.
17903
17904	@item colors
17905	Set colors separated by '|' which are going to be used for drawing of each channel.
17906
17907	@item scale
17908	Set amplitude scale.
17909
17910	Available values are:
17911	@table @samp
17912	@item lin
17913	Linear.
17914
17915	@item log
17916	Logarithmic.
17917
17918	@item sqrt
17919	Square root.
17920
17921	@item cbrt
17922	Cubic root.
17923	@end table
17924
17925	Default is linear.
17926	@end table
17927
17928	@subsection Examples
17929
17930	@itemize
17931	@item
17932	Output the input file audio and the corresponding video representation
17933	at the same time:
17934	@example
17935	amovie=a.mp3,asplit[out0],showwaves[out1]
17936	@end example
17937
17938	@item
17939	Create a synthetic signal and show it with showwaves, forcing a
17940	frame rate of 30 frames per second:
17941	@example
17942	aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
17943	@end example
17944	@end itemize
17945
17946	@section showwavespic
17947
17948	Convert input audio to a single video frame, representing the samples waves.
17949
17950	The filter accepts the following options:
17951
17952	@table @option
17953	@item size, s
17954	Specify the video size for the output. For the syntax of this option, check the
17955	@ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17956	Default value is @code{600x240}.
17957
17958	@item split_channels
17959	Set if channels should be drawn separately or overlap. Default value is 0.
17960
17961	@item colors
17962	Set colors separated by '|' which are going to be used for drawing of each channel.
17963
17964	@item scale
17965	Set amplitude scale.
17966
17967	Available values are:
17968	@table @samp
17969	@item lin
17970	Linear.
17971
17972	@item log
17973	Logarithmic.
17974
17975	@item sqrt
17976	Square root.
17977
17978	@item cbrt
17979	Cubic root.
17980	@end table
17981
17982	Default is linear.
17983	@end table
17984
17985	@subsection Examples
17986
17987	@itemize
17988	@item
17989	Extract a channel split representation of the wave form of a whole audio track
17990	in a 1024x800 picture using @command{ffmpeg}:
17991	@example
17992	ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
17993	@end example
17994	@end itemize
17995
17996	@section sidedata, asidedata
17997
17998	Delete frame side data, or select frames based on it.
17999
18000	This filter accepts the following options:
18001
18002	@table @option
18003	@item mode
18004	Set mode of operation of the filter.
18005
18006	Can be one of the following:
18007
18008	@table @samp
18009	@item select
18010	Select every frame with side data of @code{type}.
18011
18012	@item delete
18013	Delete side data of @code{type}. If @code{type} is not set, delete all side
18014	data in the frame.
18015
18016	@end table
18017
18018	@item type
18019	Set side data type used with all modes. Must be set for @code{select} mode. For
18020	the list of frame side data types, refer to the @code{AVFrameSideDataType} enum
18021	in @file{libavutil/frame.h}. For example, to choose
18022	@code{AV_FRAME_DATA_PANSCAN} side data, you must specify @code{PANSCAN}.
18023
18024	@end table
18025
18026	@section spectrumsynth
18027
18028	Sythesize audio from 2 input video spectrums, first input stream represents
18029	magnitude across time and second represents phase across time.
18030	The filter will transform from frequency domain as displayed in videos back
18031	to time domain as presented in audio output.
18032
18033	This filter is primarily created for reversing processed @ref{showspectrum}
18034	filter outputs, but can synthesize sound from other spectrograms too.
18035	But in such case results are going to be poor if the phase data is not
18036	available, because in such cases phase data need to be recreated, usually
18037	its just recreated from random noise.
18038	For best results use gray only output (@code{channel} color mode in
18039	@ref{showspectrum} filter) and @code{log} scale for magnitude video and
18040	@code{lin} scale for phase video. To produce phase, for 2nd video, use
18041	@code{data} option. Inputs videos should generally use @code{fullframe}
18042	slide mode as that saves resources needed for decoding video.
18043
18044	The filter accepts the following options:
18045
18046	@table @option
18047	@item sample_rate
18048	Specify sample rate of output audio, the sample rate of audio from which
18049	spectrum was generated may differ.
18050
18051	@item channels
18052	Set number of channels represented in input video spectrums.
18053
18054	@item scale
18055	Set scale which was used when generating magnitude input spectrum.
18056	Can be @code{lin} or @code{log}. Default is @code{log}.
18057
18058	@item slide
18059	Set slide which was used when generating inputs spectrums.
18060	Can be @code{replace}, @code{scroll}, @code{fullframe} or @code{rscroll}.
18061	Default is @code{fullframe}.
18062
18063	@item win_func
18064	Set window function used for resynthesis.
18065
18066	@item overlap
18067	Set window overlap. In range @code{[0, 1]}. Default is @code{1},
18068	which means optimal overlap for selected window function will be picked.
18069
18070	@item orientation
18071	Set orientation of input videos. Can be @code{vertical} or @code{horizontal}.
18072	Default is @code{vertical}.
18073	@end table
18074
18075	@subsection Examples
18076
18077	@itemize
18078	@item
18079	First create magnitude and phase videos from audio, assuming audio is stereo with 44100 sample rate,
18080	then resynthesize videos back to audio with spectrumsynth:
18081	@example
18082	ffmpeg -i input.flac -lavfi showspectrum=mode=separate:scale=log:overlap=0.875:color=channel:slide=fullframe:data=magnitude -an -c:v rawvideo magnitude.nut
18083	ffmpeg -i input.flac -lavfi showspectrum=mode=separate:scale=lin:overlap=0.875:color=channel:slide=fullframe:data=phase -an -c:v rawvideo phase.nut
18084	ffmpeg -i magnitude.nut -i phase.nut -lavfi spectrumsynth=channels=2:sample_rate=44100:win_func=hann:overlap=0.875:slide=fullframe output.flac
18085	@end example
18086	@end itemize
18087
18088	@section split, asplit
18089
18090	Split input into several identical outputs.
18091
18092	@code{asplit} works with audio input, @code{split} with video.
18093
18094	The filter accepts a single parameter which specifies the number of outputs. If
18095	unspecified, it defaults to 2.
18096
18097	@subsection Examples
18098
18099	@itemize
18100	@item
18101	Create two separate outputs from the same input:
18102	@example
18103	[in] split [out0][out1]
18104	@end example
18105
18106	@item
18107	To create 3 or more outputs, you need to specify the number of
18108	outputs, like in:
18109	@example
18110	[in] asplit=3 [out0][out1][out2]
18111	@end example
18112
18113	@item
18114	Create two separate outputs from the same input, one cropped and
18115	one padded:
18116	@example
18117	[in] split [splitout1][splitout2];
18118	[splitout1] crop=100:100:0:0 [cropout];
18119	[splitout2] pad=200:200:100:100 [padout];
18120	@end example
18121
18122	@item
18123	Create 5 copies of the input audio with @command{ffmpeg}:
18124	@example
18125	ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
18126	@end example
18127	@end itemize
18128
18129	@section zmq, azmq
18130
18131	Receive commands sent through a libzmq client, and forward them to
18132	filters in the filtergraph.
18133
18134	@code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
18135	must be inserted between two video filters, @code{azmq} between two
18136	audio filters.
18137
18138	To enable these filters you need to install the libzmq library and
18139	headers and configure FFmpeg with @code{--enable-libzmq}.
18140
18141	For more information about libzmq see:
18142	@url{http://www.zeromq.org/}
18143
18144	The @code{zmq} and @code{azmq} filters work as a libzmq server, which
18145	receives messages sent through a network interface defined by the
18146	@option{bind_address} option.
18147
18148	The received message must be in the form:
18149	@example
18150	@var{TARGET} @var{COMMAND} [@var{ARG}]
18151	@end example
18152
18153	@var{TARGET} specifies the target of the command, usually the name of
18154	the filter class or a specific filter instance name.
18155
18156	@var{COMMAND} specifies the name of the command for the target filter.
18157
18158	@var{ARG} is optional and specifies the optional argument list for the
18159	given @var{COMMAND}.
18160
18161	Upon reception, the message is processed and the corresponding command
18162	is injected into the filtergraph. Depending on the result, the filter
18163	will send a reply to the client, adopting the format:
18164	@example
18165	@var{ERROR_CODE} @var{ERROR_REASON}
18166	@var{MESSAGE}
18167	@end example
18168
18169	@var{MESSAGE} is optional.
18170
18171	@subsection Examples
18172
18173	Look at @file{tools/zmqsend} for an example of a zmq client which can
18174	be used to send commands processed by these filters.
18175
18176	Consider the following filtergraph generated by @command{ffplay}
18177	@example
18178	ffplay -dumpgraph 1 -f lavfi "
18179	color=s=100x100:c=red [l];
18180	color=s=100x100:c=blue [r];
18181	nullsrc=s=200x100, zmq [bg];
18182	[bg][l] overlay [bg+l];
18183	[bg+l][r] overlay=x=100 "
18184	@end example
18185
18186	To change the color of the left side of the video, the following
18187	command can be used:
18188	@example
18189	echo Parsed_color_0 c yellow | tools/zmqsend
18190	@end example
18191
18192	To change the right side:
18193	@example
18194	echo Parsed_color_1 c pink | tools/zmqsend
18195	@end example
18196
18197	@c man end MULTIMEDIA FILTERS
18198
18199	@chapter Multimedia Sources
18200	@c man begin MULTIMEDIA SOURCES
18201
18202	Below is a description of the currently available multimedia sources.
18203
18204	@section amovie
18205
18206	This is the same as @ref{movie} source, except it selects an audio
18207	stream by default.
18208
18209	@anchor{movie}
18210	@section movie
18211
18212	Read audio and/or video stream(s) from a movie container.
18213
18214	It accepts the following parameters:
18215
18216	@table @option
18217	@item filename
18218	The name of the resource to read (not necessarily a file; it can also be a
18219	device or a stream accessed through some protocol).
18220
18221	@item format_name, f
18222	Specifies the format assumed for the movie to read, and can be either
18223	the name of a container or an input device. If not specified, the
18224	format is guessed from @var{movie_name} or by probing.
18225
18226	@item seek_point, sp
18227	Specifies the seek point in seconds. The frames will be output
18228	starting from this seek point. The parameter is evaluated with
18229	@code{av_strtod}, so the numerical value may be suffixed by an IS
18230	postfix. The default value is "0".
18231
18232	@item streams, s
18233	Specifies the streams to read. Several streams can be specified,
18234	separated by "+". The source will then have as many outputs, in the
18235	same order. The syntax is explained in the ``Stream specifiers''
18236	section in the ffmpeg manual. Two special names, "dv" and "da" specify
18237	respectively the default (best suited) video and audio stream. Default
18238	is "dv", or "da" if the filter is called as "amovie".
18239
18240	@item stream_index, si
18241	Specifies the index of the video stream to read. If the value is -1,
18242	the most suitable video stream will be automatically selected. The default
18243	value is "-1". Deprecated. If the filter is called "amovie", it will select
18244	audio instead of video.
18245
18246	@item loop
18247	Specifies how many times to read the stream in sequence.
18248	If the value is 0, the stream will be looped infinitely.
18249	Default value is "1".
18250
18251	Note that when the movie is looped the source timestamps are not
18252	changed, so it will generate non monotonically increasing timestamps.
18253
18254	@item discontinuity
18255	Specifies the time difference between frames above which the point is
18256	considered a timestamp discontinuity which is removed by adjusting the later
18257	timestamps.
18258	@end table
18259
18260	It allows overlaying a second video on top of the main input of
18261	a filtergraph, as shown in this graph:
18262	@example
18263	input -----------> deltapts0 --> overlay --> output
18264	^
18265	|
18266	movie --> scale--> deltapts1 -------+
18267	@end example
18268	@subsection Examples
18269
18270	@itemize
18271	@item
18272	Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
18273	on top of the input labelled "in":
18274	@example
18275	movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
18276	[in] setpts=PTS-STARTPTS [main];
18277	[main][over] overlay=16:16 [out]
18278	@end example
18279
18280	@item
18281	Read from a video4linux2 device, and overlay it on top of the input
18282	labelled "in":
18283	@example
18284	movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
18285	[in] setpts=PTS-STARTPTS [main];
18286	[main][over] overlay=16:16 [out]
18287	@end example
18288
18289	@item
18290	Read the first video stream and the audio stream with id 0x81 from
18291	dvd.vob; the video is connected to the pad named "video" and the audio is
18292	connected to the pad named "audio":
18293	@example
18294	movie=dvd.vob:s=v:0+#0x81 [video] [audio]
18295	@end example
18296	@end itemize
18297
18298	@subsection Commands
18299
18300	Both movie and amovie support the following commands:
18301	@table @option
18302	@item seek
18303	Perform seek using "av_seek_frame".
18304	The syntax is: seek @var{stream_index}|@var{timestamp}|@var{flags}
18305	@itemize
18306	@item
18307	@var{stream_index}: If stream_index is -1, a default
18308	stream is selected, and @var{timestamp} is automatically converted
18309	from AV_TIME_BASE units to the stream specific time_base.
18310	@item
18311	@var{timestamp}: Timestamp in AVStream.time_base units
18312	or, if no stream is specified, in AV_TIME_BASE units.
18313	@item
18314	@var{flags}: Flags which select direction and seeking mode.
18315	@end itemize
18316
18317	@item get_duration
18318	Get movie duration in AV_TIME_BASE units.
18319
18320	@end table
18321
18322	@c man end MULTIMEDIA SOURCES
18323