1 @chapter Filtering Introduction
2 @c man begin FILTERING INTRODUCTION
4 Filtering in FFmpeg is enabled through the libavfilter library.
6 In libavfilter, a filter can have multiple inputs and multiple
8 To illustrate the sorts of things that are possible, we consider the
13 input --> split ---------------------> overlay --> output
16 +-----> crop --> vflip -------+
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:
25 ffmpeg -i INPUT -vf "split [main][tmp]; [tmp] crop=iw:ih/2:0:0, vflip [flip]; [main][flip] overlay=0:H/2" OUTPUT
28 The result will be that the top half of the video is mirrored
29 onto the bottom half of the output video.
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]}.
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.
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
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
54 @c man end FILTERING INTRODUCTION
57 @c man begin GRAPH2DOT
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.
68 to see how to use @file{graph2dot}.
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
74 For example the sequence of commands:
76 echo @var{GRAPH_DESCRIPTION} | \
77 tools/graph2dot -o graph.tmp && \
78 dot -Tpng graph.tmp -o graph.png && \
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:
87 ffmpeg -i infile -vf scale=640:360 outfile
89 your @var{GRAPH_DESCRIPTION} string will need to be of the form:
91 nullsrc,scale=640:360,nullsink
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.
98 @chapter Filtergraph description
99 @c man begin FILTERGRAPH DESCRIPTION
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.
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.
111 A filter with no input pads is called a "source", and a filter with no
112 output pads is called a "sink".
114 @anchor{Filtergraph syntax}
115 @section Filtergraph syntax
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}.
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.
128 A filtergraph consists of a sequence of filterchains. A sequence of
129 filterchains is represented by a list of ";"-separated filterchain
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}]
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
141 @var{arguments} is a string which contains the parameters used to
142 initialize the filter instance. It may have one of two forms:
146 A ':'-separated list of @var{key=value} pairs.
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}.
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.
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
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.
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.
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
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
190 nullsrc, split[L1], [L2]overlay, nullsink
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.
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.
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.
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.
212 Here is a BNF description of the filtergraph syntax:
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}]
223 @section Notes on filtergraph escaping
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.
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{\'}.
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.
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.
242 For example, consider the following string to be embedded in
243 the @ref{drawtext} filter description @option{text} value:
245 this is a 'string': may contain one, or more, special characters
248 This string contains the @code{'} special escaping character, and the
249 @code{:} special character, so it needs to be escaped in this way:
251 text=this is a \'string\'\: may contain one, or more, special characters
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:
258 drawtext=text=this is a \\\'string\\\'\\: may contain one\, or more\, special characters
260 (note that in addition to the @code{\'} escaping special characters,
261 also @code{,} needs to be escaped).
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:
269 -vf "drawtext=text=this is a \\\\\\'string\\\\\\'\\\\: may contain one\\, or more\\, special characters"
272 @chapter Timeline editing
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.
280 The expression accepts the following values:
283 timestamp expressed in seconds, NAN if the input timestamp is unknown
286 sequential number of the input frame, starting from 0
289 the position in the file of the input frame, NAN if unknown
293 width and height of the input frame if video
296 Additionally, these filters support an @option{enable} command that can be used
297 to re-define the expression.
299 Like any other filtering option, the @option{enable} option follows the same
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:
305 smartblur = enable='between(t,10,3*60)',
306 curves = enable='gte(t,3)' : preset=cross_process
309 See @code{ffmpeg -filters} to view which filters have timeline support.
311 @c man end FILTERGRAPH DESCRIPTION
313 @chapter Audio Filters
314 @c man begin AUDIO FILTERS
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
321 Below is a description of the currently available audio filters.
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.
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.
352 The filter accepts the following options:
356 Set input gain. Default is 1. Range is between 0.015625 and 64.
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.
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.
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.
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.
377 Set the amount by how much signal will be amplified after processing.
378 Default is 2. Range is from 1 and 64.
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.
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}.
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.
394 How much to use compressed signal in output. Default is 1.
395 Range is between 0 and 1.
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.
403 The filter accepts the following options:
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.
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}.
419 Should first stream end overlap with second stream start. Default is enabled.
422 Set curve for cross fade transition for first stream.
425 Set curve for cross fade transition for second stream.
427 For description of available curve types see @ref{afade} filter description.
434 Cross fade from one input to another:
436 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:c1=exp:c2=exp output.flac
440 Cross fade from one input to another but without overlapping:
442 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:o=0:c1=exp:c2=exp output.flac
448 Reduce audio bit resolution.
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
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.
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.
467 The filter accepts the following options:
483 Can be linear: @code{lin} or logarithmic: @code{log}.
492 Set sample reduction.
495 Enable LFO. By default disabled.
506 Delay one or more audio channels.
508 Samples in delayed channel are filled with silence.
510 The filter accepts the following option:
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.
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.
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.
541 Apply echoing to the input audio.
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.
551 A description of the accepted parameters follows.
555 Set input gain of reflected signal. Default is @code{0.6}.
558 Set output gain of reflected signal. Default is @code{0.3}.
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}.
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}.
575 Make it sound as if there are twice as many instruments as are actually playing:
577 aecho=0.8:0.88:60:0.4
581 If delay is very short, then it sound like a (metallic) robot playing music:
587 A longer delay will sound like an open air concert in the mountains:
589 aecho=0.8:0.9:1000:0.3
593 Same as above but with one more mountain:
595 aecho=0.8:0.9:1000|1800:0.3|0.25
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.
607 The filter accepts the following options:
617 Set filter mode. For restoring material use @code{reproduction} mode, otherwise
618 use @code{production} mode. Default is @code{reproduction} mode.
621 Set filter type. Selects medium. Can be one of the following:
633 select Compact Disc (CD).
639 select 50µs (FM-KF).
641 select 75µs (FM-KF).
647 Modify an audio signal according to the specified expressions.
649 This filter accepts one or more expressions (one for each channel),
650 which are evaluated and used to modify a corresponding audio signal.
652 It accepts the following parameters:
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
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.
667 Each expression in @var{exprs} can contain the following constants and functions:
671 channel number of the current expression
674 number of the evaluated sample, starting from 0
680 time of the evaluated sample expressed in seconds
683 @item nb_out_channels
684 input and output number of channels
687 the value of input channel with number @var{CH}
690 Note: this filter is slow. For faster processing you should use a
699 aeval=val(ch)/2:c=same
703 Invert phase of the second channel:
712 Apply fade-in/out effect to input audio.
714 A description of the accepted parameters follows.
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}.
721 @item start_sample, ss
722 Specify the number of the start sample for starting to apply the fade
723 effect. Default is 0.
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.
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}.
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}.
749 Set curve for fade transition.
751 It accepts the following values:
754 select triangular, linear slope (default)
756 select quarter of sine wave
758 select half of sine wave
760 select exponential sine wave
764 select inverted parabola
778 select inverted quarter of sine wave
780 select inverted half of sine wave
782 select double-exponential seat
784 select double-exponential sigmoid
792 Fade in first 15 seconds of audio:
798 Fade out last 25 seconds of a 900 seconds audio:
800 afade=t=out:st=875:d=25
805 Apply arbitrary expressions to samples in frequency domain.
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
816 Set frequency domain imaginary expression for each separate channel
817 separated by '|'. If not set, @var{real} option is used.
819 Each expression in @var{real} and @var{imag} can contain the following
827 current frequency bin number
830 number of available bins
833 channel number of the current expression
845 It accepts the following values:
861 Default is @code{w4096}
864 Set window function. Default is @code{hann}.
867 Set window overlap. If set to 1, the recommended overlap for selected
868 window function will be picked. Default is @code{0.75}.
875 Leave almost only low frequencies in audio:
877 afftfilt="1-clip((b/nb)*b,0,1)"
884 Set output format constraints for the input audio. The framework will
885 negotiate the most appropriate format to minimize conversions.
887 It accepts the following parameters:
891 A '|'-separated list of requested sample formats.
894 A '|'-separated list of requested sample rates.
896 @item channel_layouts
897 A '|'-separated list of requested channel layouts.
899 See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
900 for the required syntax.
903 If a parameter is omitted, all values are allowed.
905 Force the output to either unsigned 8-bit or signed 16-bit stereo
907 aformat=sample_fmts=u8|s16:channel_layouts=stereo
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.
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}.
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.
928 Set input level before filtering.
929 Default is 1. Allowed range is from 0.015625 to 64.
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.
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.
940 Set a ratio by which the signal is reduced.
941 Default is 2. Allowed range is from 1 to 9000.
944 Amount of milliseconds the signal has to rise above the threshold before gain
946 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
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.
954 Set amount of amplification of signal after processing.
955 Default is 1. Allowed range is from 1 to 64.
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.
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}.
966 Choose if the average level between all channels or the louder channel affects
968 Default is @code{average}. Can be @code{average} or @code{maximum}.
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.
978 The filter accepts the following options:
982 Set input gain. Default is 1.
985 Set output gain. Default is 1.
988 Don't let signals above this level pass the limiter. Default is 1.
991 The limiter will reach its attenuation level in this amount of time in
992 milliseconds. Default is 5 milliseconds.
995 Come back from limiting to attenuation 1.0 in this amount of milliseconds.
996 Default is 50 milliseconds.
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
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.
1008 Auto level output signal. Default is enabled.
1009 This normalizes audio back to 0dB if enabled.
1012 Depending on picked setting it is recommended to upsample input 2x or 4x times
1013 with @ref{aresample} before applying this filter.
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.
1022 The filter accepts the following options:
1026 Set frequency in Hz.
1029 Set method to specify band-width of filter.
1042 Specify the band-width of a filter in width_type units.
1049 The filter accepts the following options:
1053 Set the number of loops.
1056 Set maximal number of samples.
1059 Set first sample of loop.
1065 Merge two or more audio streams into a single multi-channel stream.
1067 The filter accepts the following options:
1072 Set the number of inputs. Default is 2.
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
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).
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.
1093 All inputs must have the same sample rate, and format.
1095 If inputs do not have the same duration, the output will stop with the
1098 @subsection Examples
1102 Merge two mono files into a stereo stream:
1104 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
1108 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
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
1116 Mixes multiple audio inputs into a single output.
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.
1125 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
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.
1130 It accepts the following parameters:
1134 The number of inputs. If unspecified, it defaults to 2.
1137 How to determine the end-of-stream.
1141 The duration of the longest input. (default)
1144 The duration of the shortest input.
1147 The duration of the first input.
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.
1157 @section anequalizer
1159 High-order parametric multiband equalizer for each channel.
1161 It accepts the following parameters:
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 '|'.
1171 Set channel number to which equalization will be applied.
1172 If input doesn't have that channel the entry is ignored.
1175 Set central frequency for band.
1176 If input doesn't have that frequency the entry is ignored.
1179 Set band width in hertz.
1182 Set band gain in dB.
1185 Set filter type for band, optional, can be:
1189 Butterworth, this is default.
1200 With this option activated frequency response of anequalizer is displayed
1204 Set video stream size. Only useful if curves option is activated.
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.
1213 Set frequency scale used to draw frequency response in video output.
1214 Can be linear or logarithmic. Default is logarithmic.
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.
1222 @subsection Examples
1226 Lower gain by 10 of central frequency 200Hz and width 100 Hz
1227 for first 2 channels using Chebyshev type 1 filter:
1229 anequalizer=c0 f=200 w=100 g=-10 t=1|c1 f=200 w=100 g=-10 t=1
1233 @subsection Commands
1235 This filter supports the following commands:
1238 Alter existing filter parameters.
1239 Syntax for the commands is : "@var{fN}|f=@var{freq}|w=@var{width}|g=@var{gain}"
1241 @var{fN} is existing filter number, starting from 0, if no such filter is available
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.
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=...
1253 Pass the audio source unchanged to the output.
1257 Pad the end of an audio stream with silence.
1259 This can be used together with @command{ffmpeg} @option{-shortest} to
1260 extend audio streams to the same length as the video stream.
1262 A description of the accepted options follows.
1266 Set silence packet size. Default value is 4096.
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}.
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}.
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
1284 @subsection Examples
1288 Add 1024 samples of silence to the end of the input:
1294 Make sure the audio output will contain at least 10000 samples, pad
1295 the input with silence if required:
1297 apad=whole_len=10000
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}
1306 ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
1311 Add a phasing effect to the input audio.
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.
1316 A description of the accepted parameters follows.
1320 Set input gain. Default is 0.4.
1323 Set output gain. Default is 0.74
1326 Set delay in milliseconds. Default is 3.0.
1329 Set decay. Default is 0.4.
1332 Set modulation speed in Hz. Default is 0.5.
1335 Set modulation type. Default is triangular.
1337 It accepts the following values:
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.
1360 The filter accepts the following options:
1364 Set input gain. By default it is 1. Range is [0.015625 - 64].
1367 Set output gain. By default it is 1. Range is [0.015625 - 64].
1370 Set waveform shape the LFO will use. Can be one of: sine, triangle, square,
1371 sawup or sawdown. Default is sine.
1374 Set modulation. Define how much of original signal is affected by the LFO.
1377 Set left channel offset. Default is 0. Allowed range is [0 - 1].
1380 Set right channel offset. Default is 0.5. Allowed range is [0 - 1].
1383 Set pulse width. Default is 1. Allowed range is [0 - 2].
1386 Set possible timing mode. Can be one of: bpm, ms or hz. Default is hz.
1389 Set bpm. Default is 120. Allowed range is [30 - 300]. Only used if timing
1393 Set ms. Default is 500. Allowed range is [10 - 2000]. Only used if timing
1397 Set frequency in Hz. Default is 2. Allowed range is [0.01 - 100]. Only used
1398 if timing is set to hz.
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.
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.
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.
1420 @subsection Examples
1424 Resample the input audio to 44100Hz:
1430 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
1431 samples per second compensation:
1433 aresample=async=1000
1439 Reverse an audio clip.
1441 Warning: This filter requires memory to buffer the entire clip, so trimming
1444 @subsection Examples
1448 Take the first 5 seconds of a clip, and reverse it.
1450 atrim=end=5,areverse
1454 @section asetnsamples
1456 Set the number of samples per each output audio frame.
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
1462 The filter accepts the following options:
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.
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.
1477 For example, to set the number of per-frame samples to 1234 and
1478 disable padding for the last frame, use:
1480 asetnsamples=n=1234:p=0
1485 Set the sample rate without altering the PCM data.
1486 This will result in a change of speed and pitch.
1488 The filter accepts the following options:
1491 @item sample_rate, r
1492 Set the output sample rate. Default is 44100 Hz.
1497 Show a line containing various information for each input audio frame.
1498 The input audio is not modified.
1500 The shown line contains a sequence of key/value pairs of the form
1501 @var{key}:@var{value}.
1503 The following values are shown in the output:
1507 The (sequential) number of the input frame, starting from 0.
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}.
1514 The presentation timestamp of the input frame in seconds.
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)
1527 The sample rate for the audio frame.
1530 The number of samples (per channel) in the frame.
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.
1536 @item plane_checksums
1537 A list of Adler-32 checksums for each data plane.
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.
1547 It accepts the following option:
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]}.
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
1559 Available keys for each channel are:
1590 For example full key look like this @code{lavfi.astats.1.DC_offset} or
1591 this @code{lavfi.astats.Overall.Peak_count}.
1593 For description what each key means read below.
1596 Set number of frame after which stats are going to be recalculated.
1597 Default is disabled.
1600 A description of each shown parameter follows:
1604 Mean amplitude displacement from zero.
1607 Minimal sample level.
1610 Maximal sample level.
1612 @item Min difference
1613 Minimal difference between two consecutive samples.
1615 @item Max difference
1616 Maximal difference between two consecutive samples.
1618 @item Mean difference
1619 Mean difference between two consecutive samples.
1620 The average of each difference between two consecutive samples.
1624 Standard peak and RMS level measured in dBFS.
1628 Peak and trough values for RMS level measured over a short window.
1631 Standard ratio of peak to RMS level (note: not in dB).
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}).
1638 Number of occasions (not the number of samples) that the signal attained either
1639 @var{Min level} or @var{Max level}.
1642 Overall bit depth of audio. Number of bits used for each sample.
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.
1653 @subsection Examples
1657 Slow down audio to 80% tempo:
1663 To speed up audio to 125% tempo:
1671 Trim the input so that the output contains one continuous subpart of the input.
1673 It accepts the following parameters:
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.
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.
1685 Same as @var{start}, except this option sets the start timestamp in samples
1689 Same as @var{end}, except this option sets the end timestamp in samples instead
1693 The maximum duration of the output in seconds.
1696 The number of the first sample that should be output.
1699 The number of the first sample that should be dropped.
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}.
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
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
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.
1725 Drop everything except the second minute of input:
1727 ffmpeg -i INPUT -af atrim=60:120
1731 Keep only the first 1000 samples:
1733 ffmpeg -i INPUT -af atrim=end_sample=1000
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).
1746 The filter accepts the following options:
1750 Set the filter's central frequency. Default is @code{3000}.
1753 Constant skirt gain if set to 1. Defaults to 0.
1756 Set method to specify band-width of filter.
1769 Specify the band-width of a filter in width_type units.
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).
1778 The filter accepts the following options:
1782 Set the filter's central frequency. Default is @code{3000}.
1785 Set method to specify band-width of filter.
1798 Specify the band-width of a filter in width_type units.
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).
1807 The filter accepts the following options:
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.
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.
1821 Set method to specify band-width of filter.
1834 Determine how steep is the filter's shelf transition.
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.
1844 Bauer stereo to binaural transformation, which improves headphone listening of
1845 stereo audio records.
1847 It accepts the following parameters:
1851 Pre-defined crossfeed level.
1855 Default level (fcut=700, feed=50).
1858 Chu Moy circuit (fcut=700, feed=60).
1861 Jan Meier circuit (fcut=650, feed=95).
1866 Cut frequency (in Hz).
1875 Remap input channels to new locations.
1877 It accepts the following parameters:
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.
1888 @item channel_layout
1889 The channel layout of the output stream.
1892 If no mapping is present, the filter will implicitly map input channels to
1893 output channels, preserving indices.
1895 For example, assuming a 5.1+downmix input MOV file,
1897 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1899 will create an output WAV file tagged as stereo from the downmix channels of
1902 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1904 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:5.1' out.wav
1907 @section channelsplit
1909 Split each channel from an input audio stream into a separate output stream.
1911 It accepts the following parameters:
1913 @item channel_layout
1914 The channel layout of the input stream. The default is "stereo".
1917 For example, assuming a stereo input MP3 file,
1919 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1921 will create an output Matroska file with two audio streams, one containing only
1922 the left channel and the other the right channel.
1924 Split a 5.1 WAV file into per-channel files:
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]'
1934 Add a chorus effect to the audio.
1936 Can make a single vocal sound like a chorus, but can also be applied to instrumentation.
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
1945 It accepts the following parameters:
1948 Set input gain. Default is 0.4.
1951 Set output gain. Default is 0.4.
1954 Set delays. A typical delay is around 40ms to 60ms.
1966 @subsection Examples
1972 chorus=0.7:0.9:55:0.4:0.25:2
1978 chorus=0.6:0.9:50|60:0.4|0.32:0.25|0.4:2|1.3
1982 Fuller sounding chorus with three delays:
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
1989 Compress or expand the audio's dynamic range.
1991 It accepts the following parameters:
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.
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 ....}
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}.
2019 Set the curve radius in dB for all joints. It defaults to 0.01.
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.
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.
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.
2041 @subsection Examples
2045 Make music with both quiet and loud passages suitable for listening to in a
2048 compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
2051 Another example for audio with whisper and explosion parts:
2053 compand=0|0:1|1:-90/-900|-70/-70|-30/-9|0/-3:6:0:0:0
2057 A noise gate for when the noise is at a lower level than the signal:
2059 compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
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):
2066 compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
2070 2:1 compression starting at -6dB:
2072 compand=points=-80/-80|-6/-6|0/-3.8|20/3.5
2076 2:1 compression starting at -9dB:
2078 compand=points=-80/-80|-9/-9|0/-5.3|20/2.9
2082 2:1 compression starting at -12dB:
2084 compand=points=-80/-80|-12/-12|0/-6.8|20/1.9
2088 2:1 compression starting at -18dB:
2090 compand=points=-80/-80|-18/-18|0/-9.8|20/0.7
2094 3:1 compression starting at -15dB:
2096 compand=points=-80/-80|-15/-15|0/-10.8|20/-5.2
2102 compand=points=-80/-105|-62/-80|-15.4/-15.4|0/-12|20/-7.6
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
2112 Hard limiter at -6dB:
2114 compand=attacks=0:points=-80/-80|-6/-6|20/-6
2118 Hard limiter at -12dB:
2120 compand=attacks=0:points=-80/-80|-12/-12|20/-12
2124 Hard noise gate at -35 dB:
2126 compand=attacks=0:points=-80/-115|-35.1/-80|-35/-35|20/20
2132 compand=attacks=0:points=-80/-80|-12.4/-12.4|-6/-8|0/-6.8|20/-2.8
2136 @section compensationdelay
2138 Compensation Delay Line is a metric based delay to compensate differing
2139 positions of microphones or speakers.
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.
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.
2156 It accepts the following parameters:
2160 Set millimeters distance. This is compensation distance for fine tuning.
2164 Set cm distance. This is compensation distance for tightening distance setup.
2168 Set meters distance. This is compensation distance for hard distance setup.
2172 Set dry amount. Amount of unprocessed (dry) signal.
2176 Set wet amount. Amount of processed (wet) signal.
2180 Set temperature degree in Celsius. This is the temperature of the environment.
2184 @section crystalizer
2185 Simple algorithm to expand audio dynamic range.
2187 The filter accepts the following options:
2191 Sets the intensity of effect (default: 2.0). Must be in range between 0.0
2192 (unchanged sound) to 10.0 (maximum effect).
2195 Enable clipping. By default is enabled.
2199 Apply a DC shift to the audio.
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.
2208 Set the DC shift, allowed range is [-1, 1]. It indicates the amount to shift
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.
2217 Dynamic Audio Normalizer.
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.
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.
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.
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.
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
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.
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.
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.
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.
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
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.
2376 Make audio easier to listen to on headphones.
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).
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.
2392 In order to produce complex equalisation curves, this filter can
2393 be given several times, each with a different central frequency.
2395 The filter accepts the following options:
2399 Set the filter's central frequency in Hz.
2402 Set method to specify band-width of filter.
2415 Specify the band-width of a filter in width_type units.
2418 Set the required gain or attenuation in dB.
2419 Beware of clipping when using a positive gain.
2422 @subsection Examples
2425 Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
2427 equalizer=f=1000:width_type=h:width=200:g=-10
2431 Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
2433 equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
2437 @section extrastereo
2439 Linearly increases the difference between left and right channels which
2440 adds some sort of "live" effect to playback.
2442 The filter accepts the following options:
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.
2451 Enable clipping. By default is enabled.
2454 @section firequalizer
2455 Apply FIR Equalization using arbitrary frequency response.
2457 The filter accepts the following option:
2461 Set gain curve equation (in dB). The expression can contain variables:
2464 the evaluated frequency
2468 channel number, set to 0 when multichannels evaluation is disabled
2470 channel id, see libavutil/channel_layout.h, set to the first channel id when
2471 multichannels evaluation is disabled
2475 channel_layout, see libavutil/channel_layout.h
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
2485 This option is also available as command. Default is @code{gain_interpolate(f)}.
2488 Set gain entry for gain_interpolate function. The expression can
2492 store gain entry at frequency f with value g
2494 This option is also available as command.
2497 Set filter delay in seconds. Higher value means more accurate.
2498 Default is @code{0.01}.
2501 Set filter accuracy in Hz. Lower value means more accurate.
2502 Default is @code{5}.
2505 Set window function. Acceptable values are:
2508 rectangular window, useful when gain curve is already smooth
2510 hann window (default)
2516 3-terms continuous 1st derivative nuttall window
2518 minimum 3-terms discontinuous nuttall window
2520 4-terms continuous 1st derivative nuttall window
2522 minimum 4-terms discontinuous nuttall (blackman-nuttall) window
2524 blackman-harris window
2530 If enabled, use fixed number of audio samples. This improves speed when
2531 filtering with large delay. Default is disabled.
2534 Enable multichannels evaluation on gain. Default is disabled.
2537 Enable zero phase mode by subtracting timestamp to compensate delay.
2538 Default is disabled.
2541 Set scale used by gain. Acceptable values are:
2544 linear frequency, linear gain
2546 linear frequency, logarithmic (in dB) gain (default)
2548 logarithmic (in octave scale where 20 Hz is 0) frequency, linear gain
2550 logarithmic frequency, logarithmic gain
2554 Set file for dumping, suitable for gnuplot.
2557 Set scale for dumpfile. Acceptable values are same with scale option.
2561 Enable 2-channel convolution using complex FFT. This improves speed significantly.
2562 Default is disabled.
2565 @subsection Examples
2570 firequalizer=gain='if(lt(f,1000), 0, -INF)'
2573 lowpass at 1000 Hz with gain_entry:
2575 firequalizer=gain_entry='entry(1000,0); entry(1001, -INF)'
2578 custom equalization:
2580 firequalizer=gain_entry='entry(100,0); entry(400, -4); entry(1000, -6); entry(2000, 0)'
2583 higher delay with zero phase to compensate delay:
2585 firequalizer=delay=0.1:fixed=on:zero_phase=on
2588 lowpass on left channel, highpass on right channel:
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
2596 Apply a flanging effect to the audio.
2598 The filter accepts the following options:
2602 Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
2605 Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
2608 Set percentage regeneration (delayed signal feedback). Range from -95 to 95.
2612 Set percentage of delayed signal mixed with original. Range from 0 to 100.
2613 Default value is 71.
2616 Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
2619 Set swept wave shape, can be @var{triangular} or @var{sinusoidal}.
2620 Default value is @var{sinusoidal}.
2623 Set swept wave percentage-shift for multi channel. Range from 0 to 100.
2624 Default value is 25.
2627 Set delay-line interpolation, @var{linear} or @var{quadratic}.
2628 Default is @var{linear}.
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.
2636 The filter supports the Peak Extend and Low-level Gain Adjustment features
2637 of HDCD, and detects the Transient Filter flag.
2640 ffmpeg -i HDCD16.flac -af hdcd OUT24.flac
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.
2647 ffmpeg -i HDCD16.wav -af hdcd OUT16.wav
2648 ffmpeg -i HDCD16.wav -af hdcd -acodec pcm_s24le OUT24.wav
2651 The filter accepts the following options:
2654 @item disable_autoconvert
2655 Disable any automatic format conversion or resampling in the filter graph.
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.
2662 Set the code detect timer period in ms.
2665 Always extend peaks above -3dBFS even if PE isn't signaled.
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.
2672 @code{analyze_mode=pe:force_pe=true} can be used to see all samples above the PE level.
2679 Gain adjustment level at each sample
2681 Samples where peak extend occurs
2683 Samples where the code detect timer is active
2685 Samples where the target gain does not match between channels
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).
2695 The filter accepts the following options:
2699 Set frequency in Hz. Default is 3000.
2702 Set number of poles. Default is 2.
2705 Set method to specify band-width of filter.
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.
2725 Join multiple input streams into one multi-channel stream.
2727 It accepts the following parameters:
2731 The number of input streams. It defaults to 2.
2733 @item channel_layout
2734 The desired output channel layout. It defaults to stereo.
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
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.
2749 Join 3 inputs (with properly set channel layouts):
2751 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
2754 Build a 5.1 output from 6 single-channel streams:
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'
2763 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
2765 To enable compilation of this filter you need to configure FFmpeg with
2766 @code{--enable-ladspa}.
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/}.
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.
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,
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.
2795 @item sample_rate, s
2796 Specify the sample rate, default to 44100. Only used if plugin have
2800 Set the number of samples per channel per each output frame, default
2801 is 1024. Only used if plugin have zero inputs.
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.
2815 @subsection Examples
2819 List all available plugins within amp (LADSPA example plugin) library:
2825 List all available controls and their valid ranges for @code{vcf_notch}
2826 plugin from @code{VCF} library:
2828 ladspa=f=vcf:p=vcf_notch:c=help
2832 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
2835 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
2839 Add reverberation to the audio using TAP-plugins
2840 (Tom's Audio Processing plugins):
2842 ladspa=file=tap_reverb:tap_reverb
2846 Generate white noise, with 0.2 amplitude:
2848 ladspa=file=cmt:noise_source_white:c=c0=.2
2852 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
2853 @code{C* Audio Plugin Suite} (CAPS) library:
2855 ladspa=file=caps:Click:c=c1=20'
2859 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
2861 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
2865 Increase volume by 20dB using fast lookahead limiter from Steve Harris
2866 @code{SWH Plugins} collection:
2868 ladspa=fast_lookahead_limiter_1913:fastLookaheadLimiter:20|0|2
2872 Attenuate low frequencies using Multiband EQ from Steve Harris
2873 @code{SWH Plugins} collection:
2875 ladspa=mbeq_1197:mbeq:-24|-24|-24|0|0|0|0|0|0|0|0|0|0|0|0
2879 @subsection Commands
2881 This filter supports the following commands:
2884 Modify the @var{N}-th control value.
2886 If the specified value is not valid, it is ignored and prior one is kept.
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.
2895 The filter accepts the following options:
2899 Set integrated loudness target.
2900 Range is -70.0 - -5.0. Default value is -24.0.
2903 Set loudness range target.
2904 Range is 1.0 - 20.0. Default value is 7.0.
2907 Set maximum true peak.
2908 Range is -9.0 - +0.0. Default value is -2.0.
2910 @item measured_I, measured_i
2911 Measured IL of input file.
2912 Range is -99.0 - +0.0.
2914 @item measured_LRA, measured_lra
2915 Measured LRA of input file.
2916 Range is 0.0 - 99.0.
2918 @item measured_TP, measured_tp
2919 Measured true peak of input file.
2920 Range is -99.0 - +99.0.
2922 @item measured_thresh
2923 Measured threshold of input file.
2924 Range is -99.0 - +0.0.
2927 Set offset gain. Gain is applied before the true-peak limiter.
2928 Range is -99.0 - +99.0. Default is +0.0.
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.
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.
2944 Set print format for stats. Options are summary, json, or none.
2945 Default value is none.
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).
2954 The filter accepts the following options:
2958 Set frequency in Hz. Default is 500.
2961 Set number of poles. Default is 2.
2964 Set method to specify band-width of filter.
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.
2985 Mix channels with specific gain levels. The filter accepts the output
2986 channel layout followed by a set of channels definitions.
2988 This filter is also designed to efficiently remap the channels of an audio
2991 The filter accepts parameters of the form:
2992 "@var{l}|@var{outdef}|@var{outdef}|..."
2996 output channel layout or number of channels
2999 output channel specification, of the form:
3000 "@var{out_name}=[@var{gain}*]@var{in_name}[(+-)[@var{gain}*]@var{in_name}...]"
3003 output channel to define, either a channel name (FL, FR, etc.) or a channel
3004 number (c0, c1, etc.)
3007 multiplicative coefficient for the channel, 1 leaving the volume unchanged
3010 input channel to use, see out_name for details; it is not possible to mix
3011 named and numbered input channels
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.
3018 @subsection Mixing examples
3020 For example, if you want to down-mix from stereo to mono, but with a bigger
3021 factor for the left channel:
3023 pan=1c|c0=0.9*c0+0.1*c1
3026 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
3027 7-channels surround:
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
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
3036 @subsection Remapping examples
3038 The channel remapping will be effective if, and only if:
3041 @item gain coefficients are zeroes or ones,
3042 @item only one input per channel output,
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
3049 For example, if you have a 5.1 source and want a stereo audio stream by
3050 dropping the extra channels:
3052 pan="stereo| c0=FL | c1=FR"
3055 Given the same source, you can also switch front left and front right channels
3056 and keep the input channel layout:
3058 pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
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:
3067 Still with a stereo audio stream input, you can copy the right channel in both
3068 front left and right:
3070 pan="stereo| c0=FR | c1=FR"
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}.
3081 Convert the audio sample format, sample rate and channel layout. It is
3082 not meant to be used directly.
3085 Apply time-stretching and pitch-shifting with librubberband.
3087 The filter accepts the following options:
3091 Set tempo scale factor.
3094 Set pitch scale factor.
3097 Set transients detector.
3098 Possible values are:
3107 Possible values are:
3116 Possible values are:
3123 Set processing window size.
3124 Possible values are:
3133 Possible values are:
3140 Enable formant preservation when shift pitching.
3141 Possible values are:
3149 Possible values are:
3158 Possible values are:
3165 @section sidechaincompress
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.
3174 The filter accepts the following options:
3178 Set input gain. Default is 1. Range is between 0.015625 and 64.
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.
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.
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.
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.
3199 Set the amount by how much signal will be amplified after processing.
3200 Default is 2. Range is from 1 and 64.
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.
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}.
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.
3216 Set sidechain gain. Default is 1. Range is between 0.015625 and 64.
3219 How much to use compressed signal in output. Default is 1.
3220 Range is between 0 and 1.
3223 @subsection Examples
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:
3231 ffmpeg -i main.flac -i sidechain.flac -filter_complex "[1:a]asplit=2[sc][mix];[0:a][sc]sidechaincompress[compr];[compr][mix]amerge"
3235 @section sidechaingate
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
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
3246 It needs two input streams and returns one output stream.
3247 First input stream will be processed depending on second stream signal.
3249 The filter accepts the following options:
3253 Set input level before filtering.
3254 Default is 1. Allowed range is from 0.015625 to 64.
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.
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.
3265 Set a ratio about which the signal is reduced.
3266 Default is 2. Allowed range is from 1 to 9000.
3269 Amount of milliseconds the signal has to rise above the threshold before gain
3271 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
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.
3279 Set amount of amplification of signal after processing.
3280 Default is 1. Allowed range is from 1 to 64.
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.
3287 Choose if exact signal should be taken for detection or an RMS like one.
3288 Default is rms. Can be peak or rms.
3291 Choose if the average level between all channels or the louder channel affects
3293 Default is average. Can be average or maximum.
3296 Set sidechain gain. Default is 1. Range is from 0.015625 to 64.
3299 @section silencedetect
3301 Detect silence in an audio stream.
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.
3307 The printed times and duration are expressed in seconds.
3309 The filter accepts the following options:
3313 Set silence duration until notification (default is 2 seconds).
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.
3320 @subsection Examples
3324 Detect 5 seconds of silence with -50dB noise tolerance:
3326 silencedetect=n=-50dB:d=5
3330 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
3331 tolerance in @file{silence.mp3}:
3333 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
3337 @section silenceremove
3339 Remove silence from the beginning, middle or end of the audio.
3341 The filter accepts the following options:
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}.
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}.
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}.
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}.
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
3378 Default value is @code{0}.
3380 @item stop_threshold
3381 This is the same as @option{start_threshold} but for trimming silence from
3383 Can be specified in dB (in case "dB" is appended to the specified value)
3384 or amplitude ratio. Default value is @code{0}.
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}.
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}.
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}.
3402 @subsection Examples
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:
3410 silenceremove=1:5:0.02
3414 Trim all silence encountered from beginning to end where there is more than 1
3415 second of silence in audio:
3417 silenceremove=0:0:0:-1:1:-90dB
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.
3430 To enable compilation of this filter you need to configure FFmpeg with
3431 @code{--enable-netcdf}.
3433 The filter accepts the following options:
3437 Set the SOFA file used for rendering.
3440 Set gain applied to audio. Value is in dB. Default is 0.
3443 Set rotation of virtual loudspeakers in deg. Default is 0.
3446 Set elevation of virtual speakers in deg. Default is 0.
3449 Set distance in meters between loudspeakers and the listener with near-field
3450 HRTFs. Default is 1.
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}.
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.
3469 @subsection Examples
3473 Using ClubFritz6 sofa file:
3475 sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=1
3479 Using ClubFritz12 sofa file and bigger radius with small rotation:
3481 sofalizer=sofa=/path/to/ClubFritz12.sofa:type=freq:radius=2:rotation=5
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:
3488 "sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=2:speakers=FL 45|FR 315|BL 135|BR 225:gain=28"
3492 @section stereotools
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.
3498 The filter accepts the following options:
3502 Set input level before filtering for both channels. Defaults is 1.
3503 Allowed range is from 0.015625 to 64.
3506 Set output level after filtering for both channels. Defaults is 1.
3507 Allowed range is from 0.015625 to 64.
3510 Set input balance between both channels. Default is 0.
3511 Allowed range is from -1 to 1.
3514 Set output balance between both channels. Default is 0.
3515 Allowed range is from -1 to 1.
3518 Enable softclipping. Results in analog distortion instead of harsh digital 0dB
3519 clipping. Disabled by default.
3522 Mute the left channel. Disabled by default.
3525 Mute the right channel. Disabled by default.
3528 Change the phase of the left channel. Disabled by default.
3531 Change the phase of the right channel. Disabled by default.
3534 Set stereo mode. Available values are:
3538 Left/Right to Left/Right, this is default.
3541 Left/Right to Mid/Side.
3544 Mid/Side to Left/Right.
3547 Left/Right to Left/Left.
3550 Left/Right to Right/Right.
3553 Left/Right to Left + Right.
3556 Left/Right to Right/Left.
3560 Set level of side signal. Default is 1.
3561 Allowed range is from 0.015625 to 64.
3564 Set balance of side signal. Default is 0.
3565 Allowed range is from -1 to 1.
3568 Set level of the middle signal. Default is 1.
3569 Allowed range is from 0.015625 to 64.
3572 Set middle signal pan. Default is 0. Allowed range is from -1 to 1.
3575 Set stereo base between mono and inversed channels. Default is 0.
3576 Allowed range is from -1 to 1.
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.
3583 Set S/C level. Default is 1. Allowed range is from 1 to 100.
3586 Set the stereo phase in degrees. Default is 0. Allowed range is from 0 to 360.
3589 @subsection Examples
3593 Apply karaoke like effect:
3595 stereotools=mlev=0.015625
3599 Convert M/S signal to L/R:
3601 "stereotools=mode=ms>lr"
3605 @section stereowiden
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.
3611 The filter accepts the following options:
3615 Time in milliseconds of the delay of left signal into right and vice versa.
3616 Default is 20 milliseconds.
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.
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.
3629 Set level of input signal of original channel. Default is 0.8.
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).
3638 The filter accepts the following options:
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.
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.
3652 Set method to specify band-width of filter.
3665 Determine how steep is the filter's shelf transition.
3670 Sinusoidal amplitude modulation.
3672 The filter accepts the following options:
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.
3683 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3684 Default value is 0.5.
3689 Sinusoidal phase modulation.
3691 The filter accepts the following options:
3695 Modulation frequency in Hertz.
3696 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3699 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3700 Default value is 0.5.
3705 Adjust the input audio volume.
3707 It accepts the following parameters:
3711 Set audio volume expression.
3713 Output values are clipped to the maximum value.
3715 The output audio volume is given by the relation:
3717 @var{output_volume} = @var{volume} * @var{input_volume}
3720 The default value for @var{volume} is "1.0".
3723 This parameter represents the mathematical precision.
3725 It determines which input sample formats will be allowed, which affects the
3726 precision of the volume scaling.
3730 8-bit fixed-point; this limits input sample format to U8, S16, and S32.
3732 32-bit floating-point; this limits input sample format to FLT. (default)
3734 64-bit floating-point; this limits input sample format to DBL.
3738 Choose the behaviour on encountering ReplayGain side data in input frames.
3742 Remove ReplayGain side data, ignoring its contents (the default).
3745 Ignore ReplayGain side data, but leave it in the frame.
3748 Prefer the track gain, if present.
3751 Prefer the album gain, if present.
3754 @item replaygain_preamp
3755 Pre-amplification gain in dB to apply to the selected replaygain gain.
3757 Default value for @var{replaygain_preamp} is 0.0.
3760 Set when the volume expression is evaluated.
3762 It accepts the following values:
3765 only evaluate expression once during the filter initialization, or
3766 when the @samp{volume} command is sent
3769 evaluate expression for each incoming frame
3772 Default value is @samp{once}.
3775 The volume expression can contain the following parameters.
3779 frame number (starting at zero)
3782 @item nb_consumed_samples
3783 number of samples consumed by the filter
3785 number of samples in the current frame
3787 original frame position in the file
3793 PTS at start of stream
3795 time at start of stream
3801 last set volume value
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.
3808 @subsection Commands
3810 This filter supports the following commands:
3813 Modify the volume expression.
3814 The command accepts the same syntax of the corresponding option.
3816 If the specified expression is not valid, it is kept at its current
3818 @item replaygain_noclip
3819 Prevent clipping by limiting the gain applied.
3821 Default value for @var{replaygain_noclip} is 1.
3825 @subsection Examples
3829 Halve the input audio volume:
3833 volume=volume=-6.0206dB
3836 In all the above example the named key for @option{volume} can be
3837 omitted, for example like in:
3843 Increase input audio power by 6 decibels using fixed-point precision:
3845 volume=volume=6dB:precision=fixed
3849 Fade volume after time 10 with an annihilation period of 5 seconds:
3851 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
3855 @section volumedetect
3857 Detect the volume of the input video.
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.
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
3867 All volumes are in decibels relative to the maximum PCM value.
3869 @subsection Examples
3871 Here is an excerpt of the output:
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
3887 The mean square energy is approximately -27 dB, or 10^-2.7.
3889 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
3891 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
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.
3897 @c man end AUDIO FILTERS
3899 @chapter Audio Sources
3900 @c man begin AUDIO SOURCES
3902 Below is a description of the currently available audio sources.
3906 Buffer audio frames, and make them available to the filter chain.
3908 This source is mainly intended for a programmatic use, in particular
3909 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
3911 It accepts the following parameters:
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.
3919 The sample rate of the incoming audio buffers.
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}
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}
3933 The number of channels of the incoming audio buffers.
3934 If both @var{channels} and @var{channel_layout} are specified, then they
3939 @subsection Examples
3942 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
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
3950 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
3955 Generate an audio signal specified by an expression.
3957 This source accepts in input one or more expressions (one for each
3958 channel), which are evaluated and used to generate a corresponding
3961 This source accepts the following options:
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.
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.
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
3982 If not specified, or the expressed duration is negative, the audio is
3983 supposed to be generated forever.
3986 Set the number of samples per channel per each output frame,
3989 @item sample_rate, s
3990 Specify the sample rate, default to 44100.
3993 Each expression in @var{exprs} can contain the following constants:
3997 number of the evaluated sample, starting from 0
4000 time of the evaluated sample expressed in seconds, starting from 0
4007 @subsection Examples
4017 Generate a sin signal with frequency of 440 Hz, set sample rate to
4020 aevalsrc="sin(440*2*PI*t):s=8000"
4024 Generate a two channels signal, specify the channel layout (Front
4025 Center + Back Center) explicitly:
4027 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
4031 Generate white noise:
4033 aevalsrc="-2+random(0)"
4037 Generate an amplitude modulated signal:
4039 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
4043 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
4045 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
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
4057 This source accepts the following options:
4061 @item channel_layout, cl
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}
4067 Check the channel_layout_map definition in
4068 @file{libavutil/channel_layout.c} for the mapping between strings and
4069 channel layout values.
4071 @item sample_rate, r
4072 Specifies the sample rate, and defaults to 44100.
4075 Set the number of samples per requested frames.
4079 @subsection Examples
4083 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
4085 anullsrc=r=48000:cl=4
4089 Do the same operation with a more obvious syntax:
4091 anullsrc=r=48000:cl=mono
4095 All the parameters need to be explicitly defined.
4099 Synthesize a voice utterance using the libflite library.
4101 To enable compilation of this filter you need to configure FFmpeg with
4102 @code{--enable-libflite}.
4104 Note that the flite library is not thread-safe.
4106 The filter accepts the following options:
4111 If set to 1, list the names of the available voices and exit
4112 immediately. Default value is 0.
4115 Set the maximum number of samples per frame. Default value is 512.
4118 Set the filename containing the text to speak.
4121 Set the text to speak.
4124 Set the voice to use for the speech synthesis. Default value is
4125 @code{kal}. See also the @var{list_voices} option.
4128 @subsection Examples
4132 Read from file @file{speech.txt}, and synthesize the text using the
4133 standard flite voice:
4135 flite=textfile=speech.txt
4139 Read the specified text selecting the @code{slt} voice:
4141 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4145 Input text to ffmpeg:
4147 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
4151 Make @file{ffplay} speak the specified text, using @code{flite} and
4152 the @code{lavfi} device:
4154 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
4158 For more information about libflite, check:
4159 @url{http://www.speech.cs.cmu.edu/flite/}
4163 Generate a noise audio signal.
4165 The filter accepts the following options:
4168 @item sample_rate, r
4169 Specify the sample rate. Default value is 48000 Hz.
4172 Specify the amplitude (0.0 - 1.0) of the generated audio stream. Default value
4176 Specify the duration of the generated audio stream. Not specifying this option
4177 results in noise with an infinite length.
4179 @item color, colour, c
4180 Specify the color of noise. Available noise colors are white, pink, and brown.
4181 Default color is white.
4184 Specify a value used to seed the PRNG.
4187 Set the number of samples per each output frame, default is 1024.
4190 @subsection Examples
4195 Generate 60 seconds of pink noise, with a 44.1 kHz sampling rate and an amplitude of 0.5:
4197 anoisesrc=d=60:c=pink:r=44100:a=0.5
4203 Generate an audio signal made of a sine wave with amplitude 1/8.
4205 The audio signal is bit-exact.
4207 The filter accepts the following options:
4212 Set the carrier frequency. Default is 440 Hz.
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.
4218 @item sample_rate, r
4219 Specify the sample rate, default is 44100.
4222 Specify the duration of the generated audio stream.
4224 @item samples_per_frame
4225 Set the number of samples per output frame.
4227 The expression can contain the following constants:
4231 The (sequential) number of the output audio frame, starting from 0.
4234 The PTS (Presentation TimeStamp) of the output audio frame,
4235 expressed in @var{TB} units.
4238 The PTS of the output audio frame, expressed in seconds.
4241 The timebase of the output audio frames.
4244 Default is @code{1024}.
4247 @subsection Examples
4252 Generate a simple 440 Hz sine wave:
4258 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
4262 sine=frequency=220:beep_factor=4:duration=5
4266 Generate a 1 kHz sine wave following @code{1602,1601,1602,1601,1602} NTSC
4269 sine=1000:samples_per_frame='st(0,mod(n,5)); 1602-not(not(eq(ld(0),1)+eq(ld(0),3)))'
4273 @c man end AUDIO SOURCES
4275 @chapter Audio Sinks
4276 @c man begin AUDIO SINKS
4278 Below is a description of the currently available audio sinks.
4280 @section abuffersink
4282 Buffer audio frames, and make them available to the end of filter chain.
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.
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.
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
4297 @c man end AUDIO SINKS
4299 @chapter Video Filters
4300 @c man begin VIDEO FILTERS
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
4307 Below is a description of the currently available video filters.
4309 @section alphaextract
4311 Extract the alpha component from the input as a grayscale video. This
4312 is especially useful with the @var{alphamerge} filter.
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
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:
4325 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
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.
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.
4340 This filter accepts the following option in addition to the common options from
4341 the @ref{subtitles} filter:
4345 Set the shaping engine
4347 Available values are:
4350 The default libass shaping engine, which is the best available.
4352 Fast, font-agnostic shaper that can do only substitutions
4354 Slower shaper using OpenType for substitutions and positioning
4357 The default is @code{auto}.
4361 Apply an Adaptive Temporal Averaging Denoiser to the video input.
4363 The filter accepts the following options:
4367 Set threshold A for 1st plane. Default is 0.02.
4368 Valid range is 0 to 0.3.
4371 Set threshold B for 1st plane. Default is 0.04.
4372 Valid range is 0 to 5.
4375 Set threshold A for 2nd plane. Default is 0.02.
4376 Valid range is 0 to 0.3.
4379 Set threshold B for 2nd plane. Default is 0.04.
4380 Valid range is 0 to 5.
4383 Set threshold A for 3rd plane. Default is 0.02.
4384 Valid range is 0 to 0.3.
4387 Set threshold B for 3rd plane. Default is 0.04.
4388 Valid range is 0 to 5.
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.
4394 Set number of frames filter will use for averaging. Default is 33. Must be odd
4395 number in range [5, 129].
4398 Set what planes of frame filter will use for averaging. Default is all.
4403 Apply average blur filter.
4405 The filter accepts the following options:
4409 Set horizontal kernel size.
4412 Set which planes to filter. By default all planes are filtered.
4415 Set vertical kernel size, if zero it will be same as @code{sizeX}.
4416 Default is @code{0}.
4421 Compute the bounding box for the non-black pixels in the input frame
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
4429 The filter accepts the following option:
4433 Set the minimal luminance value. Default is @code{16}.
4436 @section bitplanenoise
4438 Show and measure bit plane noise.
4440 The filter accepts the following options:
4444 Set which plane to analyze. Default is @code{1}.
4447 Filter out noisy pixels from @code{bitplane} set above.
4448 Default is disabled.
4451 @section blackdetect
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.
4458 In order to display the output lines, you need to set the loglevel at
4459 least to the AV_LOG_INFO value.
4461 The filter accepts the following options:
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.
4468 Default value is 2.0.
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:
4474 @var{nb_black_pixels} / @var{nb_pixels}
4477 for which a picture is considered black.
4478 Default value is 0.98.
4480 @item pixel_black_th, pix_th
4481 Set the threshold for considering a pixel "black".
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:
4487 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
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.
4494 Default value is 0.10.
4497 The following example sets the maximum pixel threshold to the minimum
4498 value, and detects only black intervals of 2 or more seconds:
4500 blackdetect=d=2:pix_th=0.00
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.
4510 In order to display the output lines, you need to set the loglevel at
4511 least to the AV_LOG_INFO value.
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.
4517 It accepts the following parameters:
4522 The percentage of the pixels that have to be below the threshold; it defaults to
4525 @item threshold, thresh
4526 The threshold below which a pixel value is considered black; it defaults to
4531 @section blend, tblend
4533 Blend two video frames into each other.
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.
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.
4543 A description of the accepted options follows.
4551 Set blend mode for specific pixel component or all pixel components in case
4552 of @var{all_mode}. Default value is @code{normal}.
4554 Available values for component modes are:
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.
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.
4606 The expressions can use the following variables:
4610 The sequential number of the filtered frame, starting from @code{0}.
4614 the coordinates of the current sample
4618 the width and height of currently filtered plane
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.
4628 Time of the current frame, expressed in seconds.
4631 Value of pixel component at current location for first video frame (top layer).
4634 Value of pixel component at current location for second video frame (bottom layer).
4638 Force termination when the shortest input terminates. Default is
4639 @code{0}. This option is only defined for the @code{blend} filter.
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.
4647 @subsection Examples
4651 Apply transition from bottom layer to top layer in first 10 seconds:
4653 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
4657 Apply 1x1 checkerboard effect:
4659 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
4663 Apply uncover left effect:
4665 blend=all_expr='if(gte(N*SW+X,W),A,B)'
4669 Apply uncover down effect:
4671 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
4675 Apply uncover up-left effect:
4677 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
4681 Split diagonally video and shows top and bottom layer on each side:
4683 blend=all_expr=if(gt(X,Y*(W/H)),A,B)
4687 Display differences between the current and the previous frame:
4689 tblend=all_mode=difference128
4695 Apply a boxblur algorithm to the input video.
4697 It accepts the following parameters:
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
4710 A description of the accepted options follows.
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.
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
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}.
4728 The expressions can contain the following constants:
4732 The input width and height in pixels.
4736 The input chroma image width and height in pixels.
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.
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.
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}.
4754 A value of 0 will disable the effect.
4757 @subsection Examples
4761 Apply a boxblur filter with the luma, chroma, and alpha radii
4764 boxblur=luma_radius=2:luma_power=1
4769 Set the luma radius to 2, and alpha and chroma radius to 0:
4771 boxblur=2:1:cr=0:ar=0
4775 Set the luma and chroma radii to a fraction of the video dimension:
4777 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
4783 Deinterlace the input video ("bwdif" stands for "Bob Weaver
4784 Deinterlacing Filter").
4786 Motion adaptive deinterlacing based on yadif with the use of w3fdif and cubic
4787 interpolation algorithms.
4788 It accepts the following parameters:
4792 The interlacing mode to adopt. It accepts one of the following values:
4796 Output one frame for each frame.
4798 Output one frame for each field.
4801 The default value is @code{send_field}.
4804 The picture field parity assumed for the input interlaced video. It accepts one
4805 of the following values:
4809 Assume the top field is first.
4811 Assume the bottom field is first.
4813 Enable automatic detection of field parity.
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.
4821 Specify which frames to deinterlace. Accept one of the following
4826 Deinterlace all frames.
4828 Only deinterlace frames marked as interlaced.
4831 The default value is @code{all}.
4835 YUV colorspace color/chroma keying.
4837 The filter accepts the following options:
4841 The color which will be replaced with transparency.
4844 Similarity percentage with the key color.
4846 0.01 matches only the exact key color, while 1.0 matches everything.
4851 0.0 makes pixels either fully transparent, or not transparent at all.
4853 Higher values result in semi-transparent pixels, with a higher transparency
4854 the more similar the pixels color is to the key color.
4857 Signals that the color passed is already in YUV instead of RGB.
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.
4863 @subsection Examples
4867 Make every green pixel in the input image transparent:
4869 ffmpeg -i input.png -vf chromakey=green out.png
4873 Overlay a greenscreen-video on top of a static black background.
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
4881 Display CIE color diagram with pixels overlaid onto it.
4883 The filter accepts the following options:
4898 @item uhdtv, rec2020
4911 Set what gamuts to draw.
4913 See @code{system} option for available values.
4916 Set ciescope size, by default set to 512.
4919 Set intensity used to map input pixel values to CIE diagram.
4922 Set contrast used to draw tongue colors that are out of active color system gamut.
4925 Correct gamma displayed on scope, by default enabled.
4928 Show white point on CIE diagram, by default disabled.
4931 Set input gamma. Used only with XYZ input color space.
4936 Visualize information exported by some codecs.
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.
4942 The filter accepts the following option:
4946 Set motion vectors to visualize.
4948 Available flags for @var{mv} are:
4952 forward predicted MVs of P-frames
4954 forward predicted MVs of B-frames
4956 backward predicted MVs of B-frames
4960 Display quantization parameters using the chroma planes.
4963 Set motion vectors type to visualize. Includes MVs from all frames unless specified by @var{frame_type} option.
4965 Available flags for @var{mv_type} are:
4969 forward predicted MVs
4971 backward predicted MVs
4974 @item frame_type, ft
4975 Set frame type to visualize motion vectors of.
4977 Available flags for @var{frame_type} are:
4981 intra-coded frames (I-frames)
4983 predicted frames (P-frames)
4985 bi-directionally predicted frames (B-frames)
4989 @subsection Examples
4993 Visualize forward predicted MVs of all frames using @command{ffplay}:
4995 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv_type=fp
4999 Visualize multi-directionals MVs of P and B-Frames using @command{ffplay}:
5001 ffplay -flags2 +export_mvs input.mp4 -vf codecview=mv=pf+bf+bb
5005 @section colorbalance
5006 Modify intensity of primary colors (red, green and blue) of input frames.
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.
5011 A positive adjustment value shifts the balance towards the primary color, a negative
5012 value towards the complementary color.
5014 The filter accepts the following options:
5020 Adjust red, green and blue shadows (darkest pixels).
5025 Adjust red, green and blue midtones (medium pixels).
5030 Adjust red, green and blue highlights (brightest pixels).
5032 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
5035 @subsection Examples
5039 Add red color cast to shadows:
5046 RGB colorspace color keying.
5048 The filter accepts the following options:
5052 The color which will be replaced with transparency.
5055 Similarity percentage with the key color.
5057 0.01 matches only the exact key color, while 1.0 matches everything.
5062 0.0 makes pixels either fully transparent, or not transparent at all.
5064 Higher values result in semi-transparent pixels, with a higher transparency
5065 the more similar the pixels color is to the key color.
5068 @subsection Examples
5072 Make every green pixel in the input image transparent:
5074 ffmpeg -i input.png -vf colorkey=green out.png
5078 Overlay a greenscreen-video on top of a static background image.
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
5084 @section colorlevels
5086 Adjust video input frames using levels.
5088 The filter accepts the following options:
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}.
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}.
5105 Input levels are used to lighten highlights (bright tones), darken shadows
5106 (dark tones), change the balance of bright and dark tones.
5112 Adjust red, green, blue and alpha output black point.
5113 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{0}.
5119 Adjust red, green, blue and alpha output white point.
5120 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{1}.
5122 Output levels allows manual selection of a constrained output level range.
5125 @subsection Examples
5129 Make video output darker:
5131 colorlevels=rimin=0.058:gimin=0.058:bimin=0.058
5137 colorlevels=rimin=0.039:gimin=0.039:bimin=0.039:rimax=0.96:gimax=0.96:bimax=0.96
5141 Make video output lighter:
5143 colorlevels=rimax=0.902:gimax=0.902:bimax=0.902
5147 Increase brightness:
5149 colorlevels=romin=0.5:gomin=0.5:bomin=0.5
5153 @section colorchannelmixer
5155 Adjust video input frames by re-mixing color channels.
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:
5161 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
5164 The filter accepts the following options:
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}.
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}.
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}.
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}.
5195 Allowed ranges for options are @code{[-2.0, 2.0]}.
5198 @subsection Examples
5202 Convert source to grayscale:
5204 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
5207 Simulate sepia tones:
5209 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
5213 @section colormatrix
5215 Convert color matrix.
5217 The filter accepts the following options:
5222 Specify the source and destination color matrix. Both values must be
5225 The accepted values are:
5253 For example to convert from BT.601 to SMPTE-240M, use the command:
5255 colormatrix=bt601:smpte240m
5260 Convert colorspace, transfer characteristics or color primaries.
5261 Input video needs to have an even size.
5263 The filter accepts the following options:
5268 Specify all color properties at once.
5270 The accepted values are:
5300 Specify output colorspace.
5302 The accepted values are:
5311 BT.470BG or BT.601-6 625
5314 SMPTE-170M or BT.601-6 525
5323 BT.2020 with non-constant luminance
5329 Specify output transfer characteristics.
5331 The accepted values are:
5343 Constant gamma of 2.2
5346 Constant gamma of 2.8
5349 SMPTE-170M, BT.601-6 625 or BT.601-6 525
5367 BT.2020 for 10-bits content
5370 BT.2020 for 12-bits content
5376 Specify output color primaries.
5378 The accepted values are:
5387 BT.470BG or BT.601-6 625
5390 SMPTE-170M or BT.601-6 525
5411 Specify output color range.
5413 The accepted values are:
5416 TV (restricted) range
5419 MPEG (restricted) range
5430 Specify output color format.
5432 The accepted values are:
5435 YUV 4:2:0 planar 8-bits
5438 YUV 4:2:0 planar 10-bits
5441 YUV 4:2:0 planar 12-bits
5444 YUV 4:2:2 planar 8-bits
5447 YUV 4:2:2 planar 10-bits
5450 YUV 4:2:2 planar 12-bits
5453 YUV 4:4:4 planar 8-bits
5456 YUV 4:4:4 planar 10-bits
5459 YUV 4:4:4 planar 12-bits
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.
5469 Specify dithering mode.
5471 The accepted values are:
5477 Floyd-Steinberg dithering
5481 Whitepoint adaptation mode.
5483 The accepted values are:
5486 Bradford whitepoint adaptation
5489 von Kries whitepoint adaptation
5492 identity whitepoint adaptation (i.e. no whitepoint adaptation)
5496 Override all input properties at once. Same accepted values as @ref{all}.
5499 Override input colorspace. Same accepted values as @ref{space}.
5502 Override input color primaries. Same accepted values as @ref{primaries}.
5505 Override input transfer characteristics. Same accepted values as @ref{trc}.
5508 Override input color range. Same accepted values as @ref{range}.
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.
5521 For example to convert the input to SMPTE-240M, use the command:
5523 colorspace=smpte240m
5526 @section convolution
5528 Apply convolution 3x3 or 5x5 filter.
5530 The filter accepts the following options:
5537 Set matrix for each plane.
5538 Matrix is sequence of 9 or 25 signed integers.
5544 Set multiplier for calculated value for each plane.
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.
5554 @subsection Examples
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"
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"
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"
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"
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"
5590 Copy the input source unchanged to the output. This is mainly useful for
5595 Video filtering on GPU using Apple's CoreImage API on OSX.
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
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
5606 The coreimage filter accepts the following options:
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
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
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:
5629 filter=<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...][#<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...]][#...]
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:
5636 output_rect=x\ y\ width\ height
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.
5642 output_rect=25\ 25\ 100\ 100
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
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.
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
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.
5667 @subsection Examples
5672 List all filters available:
5674 coreimage=list_filters=true
5678 Use the CIBoxBlur filter with default options to blur an image:
5680 coreimage=filter=CIBoxBlur@@default
5684 Use a filter chain with CISepiaTone at default values and CIVignetteEffect with
5685 its center at 100x100 and a radius of 50 pixels:
5687 coreimage=filter=CIBoxBlur@@default#CIVignetteEffect@@inputCenter=100\ 100@@inputRadius=50
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:
5694 ffmpeg -f lavfi -i nullsrc=s=100x100,coreimage=filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
5700 Crop the input video to given dimensions.
5702 It accepts the following parameters:
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.
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.
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.
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.
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.
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.
5736 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
5737 expressions containing the following constants:
5742 The computed values for @var{x} and @var{y}. They are evaluated for
5747 The input width and height.
5751 These are the same as @var{in_w} and @var{in_h}.
5755 The output (cropped) width and height.
5759 These are the same as @var{out_w} and @var{out_h}.
5762 same as @var{iw} / @var{ih}
5765 input sample aspect ratio
5768 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
5772 horizontal and vertical chroma subsample values. For example for the
5773 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5776 The number of the input frame, starting from 0.
5779 the position in the file of the input frame, NAN if unknown
5782 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
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}.
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.
5796 The expression for @var{x} may depend on @var{y}, and the expression
5797 for @var{y} may depend on @var{x}.
5799 @subsection Examples
5803 Crop area with size 100x100 at position (12,34).
5808 Using named options, the example above becomes:
5810 crop=w=100:h=100:x=12:y=34
5814 Crop the central input area with size 100x100:
5820 Crop the central input area with size 2/3 of the input video:
5822 crop=2/3*in_w:2/3*in_h
5826 Crop the input video central square:
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.
5837 crop=in_w-100:in_h-100:100:100
5841 Crop 10 pixels from the left and right borders, and 20 pixels from
5842 the top and bottom borders
5844 crop=in_w-2*10:in_h-2*20
5848 Keep only the bottom right quarter of the input image:
5850 crop=in_w/2:in_h/2:in_w/2:in_h/2
5854 Crop height for getting Greek harmony:
5856 crop=in_w:1/PHI*in_w
5860 Apply trembling effect:
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)
5866 Apply erratic camera effect depending on timestamp:
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)"
5872 Set x depending on the value of y:
5874 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
5878 @subsection Commands
5880 This filter supports the following commands:
5886 Set width/height of the output video and the horizontal/vertical position
5888 The command accepts the same syntax of the corresponding option.
5890 If the specified expression is not valid, it is kept at its current
5896 Auto-detect the crop size.
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.
5902 It accepts the following parameters:
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.
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.
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.
5924 This can be useful when channel logos distort the video area. 0
5925 indicates 'never reset', and returns the largest area encountered during
5932 Apply color adjustments using curves.
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
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.
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.
5951 The filter accepts the following options:
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:
5961 @item color_negative
5964 @item increase_contrast
5966 @item linear_contrast
5967 @item medium_contrast
5969 @item strong_contrast
5972 Default is @code{none}.
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.
5979 Set the key points for the red component.
5981 Set the key points for the green component.
5983 Set the key points for the blue component.
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.
5990 Specify a Photoshop curves file (@code{.acv}) to import the settings from.
5992 Save Gnuplot script of the curves in specified file.
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 ...}.
5998 @subsection Examples
6002 Increase slightly the middle level of blue:
6004 curves=blue='0/0 0.5/0.58 1/1'
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'
6012 Here we obtain the following coordinates for each components:
6015 @code{(0;0.11) (0.42;0.51) (1;0.95)}
6017 @code{(0;0) (0.50;0.48) (1;1)}
6019 @code{(0;0.22) (0.49;0.44) (1;0.80)}
6023 The previous example can also be achieved with the associated built-in preset:
6025 curves=preset=vintage
6035 Use a Photoshop preset and redefine the points of the green component:
6037 curves=psfile='MyCurvesPresets/purple.acv':green='0/0 0.45/0.53 1/1'
6041 Check out the curves of the @code{cross_process} profile using @command{ffmpeg}
6042 and @command{gnuplot}:
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
6051 Video data analysis filter.
6053 This filter shows hexadecimal pixel values of part of video.
6055 The filter accepts the following options:
6059 Set output video size.
6062 Set x offset from where to pick pixels.
6065 Set y offset from where to pick pixels.
6068 Set scope mode, can be one of the following:
6071 Draw hexadecimal pixel values with white color on black background.
6074 Draw hexadecimal pixel values with input video pixel color on black
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.
6083 Draw rows and columns numbers on left and top of video.
6086 Set background opacity.
6091 Denoise frames using 2D DCT (frequency domain filtering).
6093 This filter is not designed for real time.
6095 The filter accepts the following options:
6099 Set the noise sigma constant.
6101 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
6102 coefficient (absolute value) below this threshold with be dropped.
6104 If you need a more advanced filtering, see @option{expr}.
6106 Default is @code{0}.
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.
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.
6116 Default value is @var{blocksize}-1, which is the best possible setting.
6119 Set the coefficient factor expression.
6121 For each coefficient of a DCT block, this expression will be evaluated as a
6122 multiplier value for the coefficient.
6124 If this is option is set, the @option{sigma} option will be ignored.
6126 The absolute value of the coefficient can be accessed through the @var{c}
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.
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
6139 @subsection Examples
6141 Apply a denoise with a @option{sigma} of @code{4.5}:
6146 The same operation can be achieved using the expression system:
6148 dctdnoiz=e='gte(c, 4.5*3)'
6151 Violent denoise using a block size of @code{16x16}:
6158 Remove banding artifacts from input video.
6159 It works by replacing banded pixels with average value of referenced pixels.
6161 The filter accepts the following options:
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.
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
6177 The range defines square of four pixels around current pixel.
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
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.
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.
6201 Drop duplicated frames at regular intervals.
6203 The filter accepts the following options:
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}.
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
6217 Set scene change threshold. Default is @code{15}.
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}.
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
6234 Set whether or not chroma is considered in the metric calculations. Default is
6240 Apply deflate effect to the video.
6242 This filter replaces the pixel by the local(3x3) average by taking into account
6243 only values lower than the pixel.
6245 It accepts the following options:
6252 Limit the maximum change for each plane, default is 65535.
6253 If 0, plane will remain unchanged.
6258 Remove temporal frame luminance variations.
6260 It accepts the following options:
6264 Set moving-average filter size in frames. Default is 5. Allowed range is 2 - 129.
6267 Set averaging mode to smooth temporal luminance variations.
6269 Available values are:
6296 Remove judder produced by partially interlaced telecined content.
6298 Judder can be introduced, for instance, by @ref{pullup} filter. If the original
6299 source was partially telecined content then the output of @code{pullup,dejudder}
6300 will have a variable frame rate. May change the recorded frame rate of the
6301 container. Aside from that change, this filter will not affect constant frame
6304 The option available in this filter is:
6308 Specify the length of the window over which the judder repeats.
6310 Accepts any integer greater than 1. Useful values are:
6314 If the original was telecined from 24 to 30 fps (Film to NTSC).
6317 If the original was telecined from 25 to 30 fps (PAL to NTSC).
6320 If a mixture of the two.
6323 The default is @samp{4}.
6328 Suppress a TV station logo by a simple interpolation of the surrounding
6329 pixels. Just set a rectangle covering the logo and watch it disappear
6330 (and sometimes something even uglier appear - your mileage may vary).
6332 It accepts the following parameters:
6337 Specify the top left corner coordinates of the logo. They must be
6342 Specify the width and height of the logo to clear. They must be
6346 Specify the thickness of the fuzzy edge of the rectangle (added to
6347 @var{w} and @var{h}). The default value is 1. This option is
6348 deprecated, setting higher values should no longer be necessary and
6352 When set to 1, a green rectangle is drawn on the screen to simplify
6353 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
6354 The default value is 0.
6356 The rectangle is drawn on the outermost pixels which will be (partly)
6357 replaced with interpolated values. The values of the next pixels
6358 immediately outside this rectangle in each direction will be used to
6359 compute the interpolated pixel values inside the rectangle.
6363 @subsection Examples
6367 Set a rectangle covering the area with top left corner coordinates 0,0
6368 and size 100x77, and a band of size 10:
6370 delogo=x=0:y=0:w=100:h=77:band=10
6377 Attempt to fix small changes in horizontal and/or vertical shift. This
6378 filter helps remove camera shake from hand-holding a camera, bumping a
6379 tripod, moving on a vehicle, etc.
6381 The filter accepts the following options:
6389 Specify a rectangular area where to limit the search for motion
6391 If desired the search for motion vectors can be limited to a
6392 rectangular area of the frame defined by its top left corner, width
6393 and height. These parameters have the same meaning as the drawbox
6394 filter which can be used to visualise the position of the bounding
6397 This is useful when simultaneous movement of subjects within the frame
6398 might be confused for camera motion by the motion vector search.
6400 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
6401 then the full frame is used. This allows later options to be set
6402 without specifying the bounding box for the motion vector search.
6404 Default - search the whole frame.
6408 Specify the maximum extent of movement in x and y directions in the
6409 range 0-64 pixels. Default 16.
6412 Specify how to generate pixels to fill blanks at the edge of the
6413 frame. Available values are:
6416 Fill zeroes at blank locations
6418 Original image at blank locations
6420 Extruded edge value at blank locations
6422 Mirrored edge at blank locations
6424 Default value is @samp{mirror}.
6427 Specify the blocksize to use for motion search. Range 4-128 pixels,
6431 Specify the contrast threshold for blocks. Only blocks with more than
6432 the specified contrast (difference between darkest and lightest
6433 pixels) will be considered. Range 1-255, default 125.
6436 Specify the search strategy. Available values are:
6439 Set exhaustive search
6441 Set less exhaustive search.
6443 Default value is @samp{exhaustive}.
6446 If set then a detailed log of the motion search is written to the
6450 If set to 1, specify using OpenCL capabilities, only available if
6451 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
6457 Apply an exact inverse of the telecine operation. It requires a predefined
6458 pattern specified using the pattern option which must be the same as that passed
6459 to the telecine filter.
6461 This filter accepts the following options:
6470 The default value is @code{top}.
6474 A string of numbers representing the pulldown pattern you wish to apply.
6475 The default value is @code{23}.
6478 A number representing position of the first frame with respect to the telecine
6479 pattern. This is to be used if the stream is cut. The default value is @code{0}.
6484 Apply dilation effect to the video.
6486 This filter replaces the pixel by the local(3x3) maximum.
6488 It accepts the following options:
6495 Limit the maximum change for each plane, default is 65535.
6496 If 0, plane will remain unchanged.
6499 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
6502 Flags to local 3x3 coordinates maps like this:
6511 Displace pixels as indicated by second and third input stream.
6513 It takes three input streams and outputs one stream, the first input is the
6514 source, and second and third input are displacement maps.
6516 The second input specifies how much to displace pixels along the
6517 x-axis, while the third input specifies how much to displace pixels
6519 If one of displacement map streams terminates, last frame from that
6520 displacement map will be used.
6522 Note that once generated, displacements maps can be reused over and over again.
6524 A description of the accepted options follows.
6528 Set displace behavior for pixels that are out of range.
6530 Available values are:
6533 Missing pixels are replaced by black pixels.
6536 Adjacent pixels will spread out to replace missing pixels.
6539 Out of range pixels are wrapped so they point to pixels of other side.
6541 Default is @samp{smear}.
6545 @subsection Examples
6549 Add ripple effect to rgb input of video size hd720:
6551 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
6555 Add wave effect to rgb input of video size hd720:
6557 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
6563 Draw a colored box on the input image.
6565 It accepts the following parameters:
6570 The expressions which specify the top left corner coordinates of the box. It defaults to 0.
6574 The expressions which specify the width and height of the box; if 0 they are interpreted as
6575 the input width and height. It defaults to 0.
6578 Specify the color of the box to write. For the general syntax of this option,
6579 check the "Color" section in the ffmpeg-utils manual. If the special
6580 value @code{invert} is used, the box edge color is the same as the
6581 video with inverted luma.
6584 The expression which sets the thickness of the box edge. Default value is @code{3}.
6586 See below for the list of accepted constants.
6589 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6590 following constants:
6594 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6598 horizontal and vertical chroma subsample values. For example for the
6599 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6603 The input width and height.
6606 The input sample aspect ratio.
6610 The x and y offset coordinates where the box is drawn.
6614 The width and height of the drawn box.
6617 The thickness of the drawn box.
6619 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6620 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6624 @subsection Examples
6628 Draw a black box around the edge of the input image:
6634 Draw a box with color red and an opacity of 50%:
6636 drawbox=10:20:200:60:red@@0.5
6639 The previous example can be specified as:
6641 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
6645 Fill the box with pink color:
6647 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
6651 Draw a 2-pixel red 2.40:1 mask:
6653 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
6659 Draw a grid on the input image.
6661 It accepts the following parameters:
6666 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
6670 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
6671 input width and height, respectively, minus @code{thickness}, so image gets
6672 framed. Default to 0.
6675 Specify the color of the grid. For the general syntax of this option,
6676 check the "Color" section in the ffmpeg-utils manual. If the special
6677 value @code{invert} is used, the grid color is the same as the
6678 video with inverted luma.
6681 The expression which sets the thickness of the grid line. Default value is @code{1}.
6683 See below for the list of accepted constants.
6686 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6687 following constants:
6691 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6695 horizontal and vertical chroma subsample values. For example for the
6696 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6700 The input grid cell width and height.
6703 The input sample aspect ratio.
6707 The x and y coordinates of some point of grid intersection (meant to configure offset).
6711 The width and height of the drawn cell.
6714 The thickness of the drawn cell.
6716 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6717 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6721 @subsection Examples
6725 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
6727 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
6731 Draw a white 3x3 grid with an opacity of 50%:
6733 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
6740 Draw a text string or text from a specified file on top of a video, using the
6741 libfreetype library.
6743 To enable compilation of this filter, you need to configure FFmpeg with
6744 @code{--enable-libfreetype}.
6745 To enable default font fallback and the @var{font} option you need to
6746 configure FFmpeg with @code{--enable-libfontconfig}.
6747 To enable the @var{text_shaping} option, you need to configure FFmpeg with
6748 @code{--enable-libfribidi}.
6752 It accepts the following parameters:
6757 Used to draw a box around text using the background color.
6758 The value must be either 1 (enable) or 0 (disable).
6759 The default value of @var{box} is 0.
6762 Set the width of the border to be drawn around the box using @var{boxcolor}.
6763 The default value of @var{boxborderw} is 0.
6766 The color to be used for drawing box around text. For the syntax of this
6767 option, check the "Color" section in the ffmpeg-utils manual.
6769 The default value of @var{boxcolor} is "white".
6772 Set the line spacing in pixels of the border to be drawn around the box using @var{box}.
6773 The default value of @var{line_spacing} is 0.
6776 Set the width of the border to be drawn around the text using @var{bordercolor}.
6777 The default value of @var{borderw} is 0.
6780 Set the color to be used for drawing border around text. For the syntax of this
6781 option, check the "Color" section in the ffmpeg-utils manual.
6783 The default value of @var{bordercolor} is "black".
6786 Select how the @var{text} is expanded. Can be either @code{none},
6787 @code{strftime} (deprecated) or
6788 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
6792 Set a start time for the count. Value is in microseconds. Only applied
6793 in the deprecated strftime expansion mode. To emulate in normal expansion
6794 mode use the @code{pts} function, supplying the start time (in seconds)
6795 as the second argument.
6798 If true, check and fix text coords to avoid clipping.
6801 The color to be used for drawing fonts. For the syntax of this option, check
6802 the "Color" section in the ffmpeg-utils manual.
6804 The default value of @var{fontcolor} is "black".
6806 @item fontcolor_expr
6807 String which is expanded the same way as @var{text} to obtain dynamic
6808 @var{fontcolor} value. By default this option has empty value and is not
6809 processed. When this option is set, it overrides @var{fontcolor} option.
6812 The font family to be used for drawing text. By default Sans.
6815 The font file to be used for drawing text. The path must be included.
6816 This parameter is mandatory if the fontconfig support is disabled.
6819 Draw the text applying alpha blending. The value can
6820 be a number between 0.0 and 1.0.
6821 The expression accepts the same variables @var{x, y} as well.
6822 The default value is 1.
6823 Please see @var{fontcolor_expr}.
6826 The font size to be used for drawing text.
6827 The default value of @var{fontsize} is 16.
6830 If set to 1, attempt to shape the text (for example, reverse the order of
6831 right-to-left text and join Arabic characters) before drawing it.
6832 Otherwise, just draw the text exactly as given.
6833 By default 1 (if supported).
6836 The flags to be used for loading the fonts.
6838 The flags map the corresponding flags supported by libfreetype, and are
6839 a combination of the following values:
6846 @item vertical_layout
6847 @item force_autohint
6850 @item ignore_global_advance_width
6852 @item ignore_transform
6858 Default value is "default".
6860 For more information consult the documentation for the FT_LOAD_*
6864 The color to be used for drawing a shadow behind the drawn text. For the
6865 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
6867 The default value of @var{shadowcolor} is "black".
6871 The x and y offsets for the text shadow position with respect to the
6872 position of the text. They can be either positive or negative
6873 values. The default value for both is "0".
6876 The starting frame number for the n/frame_num variable. The default value
6880 The size in number of spaces to use for rendering the tab.
6884 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
6885 format. It can be used with or without text parameter. @var{timecode_rate}
6886 option must be specified.
6888 @item timecode_rate, rate, r
6889 Set the timecode frame rate (timecode only).
6892 If set to 1, the output of the timecode option will wrap around at 24 hours.
6893 Default is 0 (disabled).
6896 The text string to be drawn. The text must be a sequence of UTF-8
6898 This parameter is mandatory if no file is specified with the parameter
6902 A text file containing text to be drawn. The text must be a sequence
6903 of UTF-8 encoded characters.
6905 This parameter is mandatory if no text string is specified with the
6906 parameter @var{text}.
6908 If both @var{text} and @var{textfile} are specified, an error is thrown.
6911 If set to 1, the @var{textfile} will be reloaded before each frame.
6912 Be sure to update it atomically, or it may be read partially, or even fail.
6916 The expressions which specify the offsets where text will be drawn
6917 within the video frame. They are relative to the top/left border of the
6920 The default value of @var{x} and @var{y} is "0".
6922 See below for the list of accepted constants and functions.
6925 The parameters for @var{x} and @var{y} are expressions containing the
6926 following constants and functions:
6930 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
6934 horizontal and vertical chroma subsample values. For example for the
6935 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6938 the height of each text line
6946 @item max_glyph_a, ascent
6947 the maximum distance from the baseline to the highest/upper grid
6948 coordinate used to place a glyph outline point, for all the rendered
6950 It is a positive value, due to the grid's orientation with the Y axis
6953 @item max_glyph_d, descent
6954 the maximum distance from the baseline to the lowest grid coordinate
6955 used to place a glyph outline point, for all the rendered glyphs.
6956 This is a negative value, due to the grid's orientation, with the Y axis
6960 maximum glyph height, that is the maximum height for all the glyphs
6961 contained in the rendered text, it is equivalent to @var{ascent} -
6965 maximum glyph width, that is the maximum width for all the glyphs
6966 contained in the rendered text
6969 the number of input frame, starting from 0
6971 @item rand(min, max)
6972 return a random number included between @var{min} and @var{max}
6975 The input sample aspect ratio.
6978 timestamp expressed in seconds, NAN if the input timestamp is unknown
6981 the height of the rendered text
6984 the width of the rendered text
6988 the x and y offset coordinates where the text is drawn.
6990 These parameters allow the @var{x} and @var{y} expressions to refer
6991 each other, so you can for example specify @code{y=x/dar}.
6994 @anchor{drawtext_expansion}
6995 @subsection Text expansion
6997 If @option{expansion} is set to @code{strftime},
6998 the filter recognizes strftime() sequences in the provided text and
6999 expands them accordingly. Check the documentation of strftime(). This
7000 feature is deprecated.
7002 If @option{expansion} is set to @code{none}, the text is printed verbatim.
7004 If @option{expansion} is set to @code{normal} (which is the default),
7005 the following expansion mechanism is used.
7007 The backslash character @samp{\}, followed by any character, always expands to
7008 the second character.
7010 Sequences of the form @code{%@{...@}} are expanded. The text between the
7011 braces is a function name, possibly followed by arguments separated by ':'.
7012 If the arguments contain special characters or delimiters (':' or '@}'),
7013 they should be escaped.
7015 Note that they probably must also be escaped as the value for the
7016 @option{text} option in the filter argument string and as the filter
7017 argument in the filtergraph description, and possibly also for the shell,
7018 that makes up to four levels of escaping; using a text file avoids these
7021 The following functions are available:
7026 The expression evaluation result.
7028 It must take one argument specifying the expression to be evaluated,
7029 which accepts the same constants and functions as the @var{x} and
7030 @var{y} values. Note that not all constants should be used, for
7031 example the text size is not known when evaluating the expression, so
7032 the constants @var{text_w} and @var{text_h} will have an undefined
7035 @item expr_int_format, eif
7036 Evaluate the expression's value and output as formatted integer.
7038 The first argument is the expression to be evaluated, just as for the @var{expr} function.
7039 The second argument specifies the output format. Allowed values are @samp{x},
7040 @samp{X}, @samp{d} and @samp{u}. They are treated exactly as in the
7041 @code{printf} function.
7042 The third parameter is optional and sets the number of positions taken by the output.
7043 It can be used to add padding with zeros from the left.
7046 The time at which the filter is running, expressed in UTC.
7047 It can accept an argument: a strftime() format string.
7050 The time at which the filter is running, expressed in the local time zone.
7051 It can accept an argument: a strftime() format string.
7054 Frame metadata. Takes one or two arguments.
7056 The first argument is mandatory and specifies the metadata key.
7058 The second argument is optional and specifies a default value, used when the
7059 metadata key is not found or empty.
7062 The frame number, starting from 0.
7065 A 1 character description of the current picture type.
7068 The timestamp of the current frame.
7069 It can take up to three arguments.
7071 The first argument is the format of the timestamp; it defaults to @code{flt}
7072 for seconds as a decimal number with microsecond accuracy; @code{hms} stands
7073 for a formatted @var{[-]HH:MM:SS.mmm} timestamp with millisecond accuracy.
7074 @code{gmtime} stands for the timestamp of the frame formatted as UTC time;
7075 @code{localtime} stands for the timestamp of the frame formatted as
7076 local time zone time.
7078 The second argument is an offset added to the timestamp.
7080 If the format is set to @code{localtime} or @code{gmtime},
7081 a third argument may be supplied: a strftime() format string.
7082 By default, @var{YYYY-MM-DD HH:MM:SS} format will be used.
7085 @subsection Examples
7089 Draw "Test Text" with font FreeSerif, using the default values for the
7090 optional parameters.
7093 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
7097 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
7098 and y=50 (counting from the top-left corner of the screen), text is
7099 yellow with a red box around it. Both the text and the box have an
7103 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
7104 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
7107 Note that the double quotes are not necessary if spaces are not used
7108 within the parameter list.
7111 Show the text at the center of the video frame:
7113 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h)/2"
7117 Show the text at a random position, switching to a new position every 30 seconds:
7119 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)"
7123 Show a text line sliding from right to left in the last row of the video
7124 frame. The file @file{LONG_LINE} is assumed to contain a single line
7127 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
7131 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
7133 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
7137 Draw a single green letter "g", at the center of the input video.
7138 The glyph baseline is placed at half screen height.
7140 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
7144 Show text for 1 second every 3 seconds:
7146 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
7150 Use fontconfig to set the font. Note that the colons need to be escaped.
7152 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
7156 Print the date of a real-time encoding (see strftime(3)):
7158 drawtext='fontfile=FreeSans.ttf:text=%@{localtime\:%a %b %d %Y@}'
7162 Show text fading in and out (appearing/disappearing):
7165 DS=1.0 # display start
7166 DE=10.0 # display end
7167 FID=1.5 # fade in duration
7168 FOD=5 # fade out duration
7169 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 @}"
7173 Horizontally align multiple separate texts. Note that @option{max_glyph_a}
7174 and the @option{fontsize} value are included in the @option{y} offset.
7176 drawtext=fontfile=FreeSans.ttf:text=DOG:fontsize=24:x=10:y=20+24-max_glyph_a,
7177 drawtext=fontfile=FreeSans.ttf:text=cow:fontsize=24:x=80:y=20+24-max_glyph_a
7182 For more information about libfreetype, check:
7183 @url{http://www.freetype.org/}.
7185 For more information about fontconfig, check:
7186 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
7188 For more information about libfribidi, check:
7189 @url{http://fribidi.org/}.
7193 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
7195 The filter accepts the following options:
7200 Set low and high threshold values used by the Canny thresholding
7203 The high threshold selects the "strong" edge pixels, which are then
7204 connected through 8-connectivity with the "weak" edge pixels selected
7205 by the low threshold.
7207 @var{low} and @var{high} threshold values must be chosen in the range
7208 [0,1], and @var{low} should be lesser or equal to @var{high}.
7210 Default value for @var{low} is @code{20/255}, and default value for @var{high}
7214 Define the drawing mode.
7218 Draw white/gray wires on black background.
7221 Mix the colors to create a paint/cartoon effect.
7224 Default value is @var{wires}.
7227 @subsection Examples
7231 Standard edge detection with custom values for the hysteresis thresholding:
7233 edgedetect=low=0.1:high=0.4
7237 Painting effect without thresholding:
7239 edgedetect=mode=colormix:high=0
7244 Set brightness, contrast, saturation and approximate gamma adjustment.
7246 The filter accepts the following options:
7250 Set the contrast expression. The value must be a float value in range
7251 @code{-2.0} to @code{2.0}. The default value is "1".
7254 Set the brightness expression. The value must be a float value in
7255 range @code{-1.0} to @code{1.0}. The default value is "0".
7258 Set the saturation expression. The value must be a float in
7259 range @code{0.0} to @code{3.0}. The default value is "1".
7262 Set the gamma expression. The value must be a float in range
7263 @code{0.1} to @code{10.0}. The default value is "1".
7266 Set the gamma expression for red. The value must be a float in
7267 range @code{0.1} to @code{10.0}. The default value is "1".
7270 Set the gamma expression for green. The value must be a float in range
7271 @code{0.1} to @code{10.0}. The default value is "1".
7274 Set the gamma expression for blue. The value must be a float in range
7275 @code{0.1} to @code{10.0}. The default value is "1".
7278 Set the gamma weight expression. It can be used to reduce the effect
7279 of a high gamma value on bright image areas, e.g. keep them from
7280 getting overamplified and just plain white. The value must be a float
7281 in range @code{0.0} to @code{1.0}. A value of @code{0.0} turns the
7282 gamma correction all the way down while @code{1.0} leaves it at its
7283 full strength. Default is "1".
7286 Set when the expressions for brightness, contrast, saturation and
7287 gamma expressions are evaluated.
7289 It accepts the following values:
7292 only evaluate expressions once during the filter initialization or
7293 when a command is processed
7296 evaluate expressions for each incoming frame
7299 Default value is @samp{init}.
7302 The expressions accept the following parameters:
7305 frame count of the input frame starting from 0
7308 byte position of the corresponding packet in the input file, NAN if
7312 frame rate of the input video, NAN if the input frame rate is unknown
7315 timestamp expressed in seconds, NAN if the input timestamp is unknown
7318 @subsection Commands
7319 The filter supports the following commands:
7323 Set the contrast expression.
7326 Set the brightness expression.
7329 Set the saturation expression.
7332 Set the gamma expression.
7335 Set the gamma_r expression.
7338 Set gamma_g expression.
7341 Set gamma_b expression.
7344 Set gamma_weight expression.
7346 The command accepts the same syntax of the corresponding option.
7348 If the specified expression is not valid, it is kept at its current
7355 Apply erosion effect to the video.
7357 This filter replaces the pixel by the local(3x3) minimum.
7359 It accepts the following options:
7366 Limit the maximum change for each plane, default is 65535.
7367 If 0, plane will remain unchanged.
7370 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
7373 Flags to local 3x3 coordinates maps like this:
7380 @section extractplanes
7382 Extract color channel components from input video stream into
7383 separate grayscale video streams.
7385 The filter accepts the following option:
7389 Set plane(s) to extract.
7391 Available values for planes are:
7402 Choosing planes not available in the input will result in an error.
7403 That means you cannot select @code{r}, @code{g}, @code{b} planes
7404 with @code{y}, @code{u}, @code{v} planes at same time.
7407 @subsection Examples
7411 Extract luma, u and v color channel component from input video frame
7412 into 3 grayscale outputs:
7414 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
7420 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
7422 For each input image, the filter will compute the optimal mapping from
7423 the input to the output given the codebook length, that is the number
7424 of distinct output colors.
7426 This filter accepts the following options.
7429 @item codebook_length, l
7430 Set codebook length. The value must be a positive integer, and
7431 represents the number of distinct output colors. Default value is 256.
7434 Set the maximum number of iterations to apply for computing the optimal
7435 mapping. The higher the value the better the result and the higher the
7436 computation time. Default value is 1.
7439 Set a random seed, must be an integer included between 0 and
7440 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
7441 will try to use a good random seed on a best effort basis.
7444 Set pal8 output pixel format. This option does not work with codebook
7445 length greater than 256.
7450 Apply a fade-in/out effect to the input video.
7452 It accepts the following parameters:
7456 The effect type can be either "in" for a fade-in, or "out" for a fade-out
7458 Default is @code{in}.
7460 @item start_frame, s
7461 Specify the number of the frame to start applying the fade
7462 effect at. Default is 0.
7465 The number of frames that the fade effect lasts. At the end of the
7466 fade-in effect, the output video will have the same intensity as the input video.
7467 At the end of the fade-out transition, the output video will be filled with the
7468 selected @option{color}.
7472 If set to 1, fade only alpha channel, if one exists on the input.
7475 @item start_time, st
7476 Specify the timestamp (in seconds) of the frame to start to apply the fade
7477 effect. If both start_frame and start_time are specified, the fade will start at
7478 whichever comes last. Default is 0.
7481 The number of seconds for which the fade effect has to last. At the end of the
7482 fade-in effect the output video will have the same intensity as the input video,
7483 at the end of the fade-out transition the output video will be filled with the
7484 selected @option{color}.
7485 If both duration and nb_frames are specified, duration is used. Default is 0
7486 (nb_frames is used by default).
7489 Specify the color of the fade. Default is "black".
7492 @subsection Examples
7496 Fade in the first 30 frames of video:
7501 The command above is equivalent to:
7507 Fade out the last 45 frames of a 200-frame video:
7510 fade=type=out:start_frame=155:nb_frames=45
7514 Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
7516 fade=in:0:25, fade=out:975:25
7520 Make the first 5 frames yellow, then fade in from frame 5-24:
7522 fade=in:5:20:color=yellow
7526 Fade in alpha over first 25 frames of video:
7528 fade=in:0:25:alpha=1
7532 Make the first 5.5 seconds black, then fade in for 0.5 seconds:
7534 fade=t=in:st=5.5:d=0.5
7540 Apply arbitrary expressions to samples in frequency domain
7544 Adjust the dc value (gain) of the luma plane of the image. The filter
7545 accepts an integer value in range @code{0} to @code{1000}. The default
7546 value is set to @code{0}.
7549 Adjust the dc value (gain) of the 1st chroma plane of the image. The
7550 filter accepts an integer value in range @code{0} to @code{1000}. The
7551 default value is set to @code{0}.
7554 Adjust the dc value (gain) of the 2nd chroma plane of the image. The
7555 filter accepts an integer value in range @code{0} to @code{1000}. The
7556 default value is set to @code{0}.
7559 Set the frequency domain weight expression for the luma plane.
7562 Set the frequency domain weight expression for the 1st chroma plane.
7565 Set the frequency domain weight expression for the 2nd chroma plane.
7567 The filter accepts the following variables:
7570 The coordinates of the current sample.
7574 The width and height of the image.
7577 @subsection Examples
7583 fftfilt=dc_Y=128:weight_Y='squish(1-(Y+X)/100)'
7589 fftfilt=dc_Y=0:weight_Y='squish((Y+X)/100-1)'
7595 fftfilt=dc_Y=0:weight_Y='1+squish(1-(Y+X)/100)'
7601 fftfilt=dc_Y=0:weight_Y='exp(-4 * ((Y+X)/(W+H)))'
7608 Extract a single field from an interlaced image using stride
7609 arithmetic to avoid wasting CPU time. The output frames are marked as
7612 The filter accepts the following options:
7616 Specify whether to extract the top (if the value is @code{0} or
7617 @code{top}) or the bottom field (if the value is @code{1} or
7623 Create new frames by copying the top and bottom fields from surrounding frames
7624 supplied as numbers by the hint file.
7628 Set file containing hints: absolute/relative frame numbers.
7630 There must be one line for each frame in a clip. Each line must contain two
7631 numbers separated by the comma, optionally followed by @code{-} or @code{+}.
7632 Numbers supplied on each line of file can not be out of [N-1,N+1] where N
7633 is current frame number for @code{absolute} mode or out of [-1, 1] range
7634 for @code{relative} mode. First number tells from which frame to pick up top
7635 field and second number tells from which frame to pick up bottom field.
7637 If optionally followed by @code{+} output frame will be marked as interlaced,
7638 else if followed by @code{-} output frame will be marked as progressive, else
7639 it will be marked same as input frame.
7640 If line starts with @code{#} or @code{;} that line is skipped.
7643 Can be item @code{absolute} or @code{relative}. Default is @code{absolute}.
7646 Example of first several lines of @code{hint} file for @code{relative} mode:
7649 1,0 - # second frame, use third's frame top field and second's frame bottom field
7650 1,0 - # third frame, use fourth's frame top field and third's frame bottom field
7667 Field matching filter for inverse telecine. It is meant to reconstruct the
7668 progressive frames from a telecined stream. The filter does not drop duplicated
7669 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
7670 followed by a decimation filter such as @ref{decimate} in the filtergraph.
7672 The separation of the field matching and the decimation is notably motivated by
7673 the possibility of inserting a de-interlacing filter fallback between the two.
7674 If the source has mixed telecined and real interlaced content,
7675 @code{fieldmatch} will not be able to match fields for the interlaced parts.
7676 But these remaining combed frames will be marked as interlaced, and thus can be
7677 de-interlaced by a later filter such as @ref{yadif} before decimation.
7679 In addition to the various configuration options, @code{fieldmatch} can take an
7680 optional second stream, activated through the @option{ppsrc} option. If
7681 enabled, the frames reconstruction will be based on the fields and frames from
7682 this second stream. This allows the first input to be pre-processed in order to
7683 help the various algorithms of the filter, while keeping the output lossless
7684 (assuming the fields are matched properly). Typically, a field-aware denoiser,
7685 or brightness/contrast adjustments can help.
7687 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
7688 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
7689 which @code{fieldmatch} is based on. While the semantic and usage are very
7690 close, some behaviour and options names can differ.
7692 The @ref{decimate} filter currently only works for constant frame rate input.
7693 If your input has mixed telecined (30fps) and progressive content with a lower
7694 framerate like 24fps use the following filterchain to produce the necessary cfr
7695 stream: @code{dejudder,fps=30000/1001,fieldmatch,decimate}.
7697 The filter accepts the following options:
7701 Specify the assumed field order of the input stream. Available values are:
7705 Auto detect parity (use FFmpeg's internal parity value).
7707 Assume bottom field first.
7709 Assume top field first.
7712 Note that it is sometimes recommended not to trust the parity announced by the
7715 Default value is @var{auto}.
7718 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
7719 sense that it won't risk creating jerkiness due to duplicate frames when
7720 possible, but if there are bad edits or blended fields it will end up
7721 outputting combed frames when a good match might actually exist. On the other
7722 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
7723 but will almost always find a good frame if there is one. The other values are
7724 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
7725 jerkiness and creating duplicate frames versus finding good matches in sections
7726 with bad edits, orphaned fields, blended fields, etc.
7728 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
7730 Available values are:
7734 2-way matching (p/c)
7736 2-way matching, and trying 3rd match if still combed (p/c + n)
7738 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
7740 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
7741 still combed (p/c + n + u/b)
7743 3-way matching (p/c/n)
7745 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
7746 detected as combed (p/c/n + u/b)
7749 The parenthesis at the end indicate the matches that would be used for that
7750 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
7753 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
7756 Default value is @var{pc_n}.
7759 Mark the main input stream as a pre-processed input, and enable the secondary
7760 input stream as the clean source to pick the fields from. See the filter
7761 introduction for more details. It is similar to the @option{clip2} feature from
7764 Default value is @code{0} (disabled).
7767 Set the field to match from. It is recommended to set this to the same value as
7768 @option{order} unless you experience matching failures with that setting. In
7769 certain circumstances changing the field that is used to match from can have a
7770 large impact on matching performance. Available values are:
7774 Automatic (same value as @option{order}).
7776 Match from the bottom field.
7778 Match from the top field.
7781 Default value is @var{auto}.
7784 Set whether or not chroma is included during the match comparisons. In most
7785 cases it is recommended to leave this enabled. You should set this to @code{0}
7786 only if your clip has bad chroma problems such as heavy rainbowing or other
7787 artifacts. Setting this to @code{0} could also be used to speed things up at
7788 the cost of some accuracy.
7790 Default value is @code{1}.
7794 These define an exclusion band which excludes the lines between @option{y0} and
7795 @option{y1} from being included in the field matching decision. An exclusion
7796 band can be used to ignore subtitles, a logo, or other things that may
7797 interfere with the matching. @option{y0} sets the starting scan line and
7798 @option{y1} sets the ending line; all lines in between @option{y0} and
7799 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
7800 @option{y0} and @option{y1} to the same value will disable the feature.
7801 @option{y0} and @option{y1} defaults to @code{0}.
7804 Set the scene change detection threshold as a percentage of maximum change on
7805 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
7806 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
7807 @option{scthresh} is @code{[0.0, 100.0]}.
7809 Default value is @code{12.0}.
7812 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
7813 account the combed scores of matches when deciding what match to use as the
7814 final match. Available values are:
7818 No final matching based on combed scores.
7820 Combed scores are only used when a scene change is detected.
7822 Use combed scores all the time.
7825 Default is @var{sc}.
7828 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
7829 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
7830 Available values are:
7834 No forced calculation.
7836 Force p/c/n calculations.
7838 Force p/c/n/u/b calculations.
7841 Default value is @var{none}.
7844 This is the area combing threshold used for combed frame detection. This
7845 essentially controls how "strong" or "visible" combing must be to be detected.
7846 Larger values mean combing must be more visible and smaller values mean combing
7847 can be less visible or strong and still be detected. Valid settings are from
7848 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
7849 be detected as combed). This is basically a pixel difference value. A good
7850 range is @code{[8, 12]}.
7852 Default value is @code{9}.
7855 Sets whether or not chroma is considered in the combed frame decision. Only
7856 disable this if your source has chroma problems (rainbowing, etc.) that are
7857 causing problems for the combed frame detection with chroma enabled. Actually,
7858 using @option{chroma}=@var{0} is usually more reliable, except for the case
7859 where there is chroma only combing in the source.
7861 Default value is @code{0}.
7865 Respectively set the x-axis and y-axis size of the window used during combed
7866 frame detection. This has to do with the size of the area in which
7867 @option{combpel} pixels are required to be detected as combed for a frame to be
7868 declared combed. See the @option{combpel} parameter description for more info.
7869 Possible values are any number that is a power of 2 starting at 4 and going up
7872 Default value is @code{16}.
7875 The number of combed pixels inside any of the @option{blocky} by
7876 @option{blockx} size blocks on the frame for the frame to be detected as
7877 combed. While @option{cthresh} controls how "visible" the combing must be, this
7878 setting controls "how much" combing there must be in any localized area (a
7879 window defined by the @option{blockx} and @option{blocky} settings) on the
7880 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
7881 which point no frames will ever be detected as combed). This setting is known
7882 as @option{MI} in TFM/VFM vocabulary.
7884 Default value is @code{80}.
7887 @anchor{p/c/n/u/b meaning}
7888 @subsection p/c/n/u/b meaning
7890 @subsubsection p/c/n
7892 We assume the following telecined stream:
7895 Top fields: 1 2 2 3 4
7896 Bottom fields: 1 2 3 4 4
7899 The numbers correspond to the progressive frame the fields relate to. Here, the
7900 first two frames are progressive, the 3rd and 4th are combed, and so on.
7902 When @code{fieldmatch} is configured to run a matching from bottom
7903 (@option{field}=@var{bottom}) this is how this input stream get transformed:
7908 B 1 2 3 4 4 <-- matching reference
7917 As a result of the field matching, we can see that some frames get duplicated.
7918 To perform a complete inverse telecine, you need to rely on a decimation filter
7919 after this operation. See for instance the @ref{decimate} filter.
7921 The same operation now matching from top fields (@option{field}=@var{top})
7926 T 1 2 2 3 4 <-- matching reference
7936 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
7937 basically, they refer to the frame and field of the opposite parity:
7940 @item @var{p} matches the field of the opposite parity in the previous frame
7941 @item @var{c} matches the field of the opposite parity in the current frame
7942 @item @var{n} matches the field of the opposite parity in the next frame
7947 The @var{u} and @var{b} matching are a bit special in the sense that they match
7948 from the opposite parity flag. In the following examples, we assume that we are
7949 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
7950 'x' is placed above and below each matched fields.
7952 With bottom matching (@option{field}=@var{bottom}):
7957 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7958 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
7966 With top matching (@option{field}=@var{top}):
7971 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7972 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
7980 @subsection Examples
7982 Simple IVTC of a top field first telecined stream:
7984 fieldmatch=order=tff:combmatch=none, decimate
7987 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
7989 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
7994 Transform the field order of the input video.
7996 It accepts the following parameters:
8001 The output field order. Valid values are @var{tff} for top field first or @var{bff}
8002 for bottom field first.
8005 The default value is @samp{tff}.
8007 The transformation is done by shifting the picture content up or down
8008 by one line, and filling the remaining line with appropriate picture content.
8009 This method is consistent with most broadcast field order converters.
8011 If the input video is not flagged as being interlaced, or it is already
8012 flagged as being of the required output field order, then this filter does
8013 not alter the incoming video.
8015 It is very useful when converting to or from PAL DV material,
8016 which is bottom field first.
8020 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
8023 @section fifo, afifo
8025 Buffer input images and send them when they are requested.
8027 It is mainly useful when auto-inserted by the libavfilter
8030 It does not take parameters.
8034 Find a rectangular object
8036 It accepts the following options:
8040 Filepath of the object image, needs to be in gray8.
8043 Detection threshold, default is 0.5.
8046 Number of mipmaps, default is 3.
8048 @item xmin, ymin, xmax, ymax
8049 Specifies the rectangle in which to search.
8052 @subsection Examples
8056 Generate a representative palette of a given video using @command{ffmpeg}:
8058 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
8064 Cover a rectangular object
8066 It accepts the following options:
8070 Filepath of the optional cover image, needs to be in yuv420.
8075 It accepts the following values:
8078 cover it by the supplied image
8080 cover it by interpolating the surrounding pixels
8083 Default value is @var{blur}.
8086 @subsection Examples
8090 Generate a representative palette of a given video using @command{ffmpeg}:
8092 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
8099 Convert the input video to one of the specified pixel formats.
8100 Libavfilter will try to pick one that is suitable as input to
8103 It accepts the following parameters:
8107 A '|'-separated list of pixel format names, such as
8108 "pix_fmts=yuv420p|monow|rgb24".
8112 @subsection Examples
8116 Convert the input video to the @var{yuv420p} format
8118 format=pix_fmts=yuv420p
8121 Convert the input video to any of the formats in the list
8123 format=pix_fmts=yuv420p|yuv444p|yuv410p
8130 Convert the video to specified constant frame rate by duplicating or dropping
8131 frames as necessary.
8133 It accepts the following parameters:
8137 The desired output frame rate. The default is @code{25}.
8142 Possible values are:
8145 zero round towards 0
8149 round towards -infinity
8151 round towards +infinity
8155 The default is @code{near}.
8158 Assume the first PTS should be the given value, in seconds. This allows for
8159 padding/trimming at the start of stream. By default, no assumption is made
8160 about the first frame's expected PTS, so no padding or trimming is done.
8161 For example, this could be set to 0 to pad the beginning with duplicates of
8162 the first frame if a video stream starts after the audio stream or to trim any
8163 frames with a negative PTS.
8167 Alternatively, the options can be specified as a flat string:
8168 @var{fps}[:@var{round}].
8170 See also the @ref{setpts} filter.
8172 @subsection Examples
8176 A typical usage in order to set the fps to 25:
8182 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
8184 fps=fps=film:round=near
8190 Pack two different video streams into a stereoscopic video, setting proper
8191 metadata on supported codecs. The two views should have the same size and
8192 framerate and processing will stop when the shorter video ends. Please note
8193 that you may conveniently adjust view properties with the @ref{scale} and
8196 It accepts the following parameters:
8200 The desired packing format. Supported values are:
8205 The views are next to each other (default).
8208 The views are on top of each other.
8211 The views are packed by line.
8214 The views are packed by column.
8217 The views are temporally interleaved.
8226 # Convert left and right views into a frame-sequential video
8227 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
8229 # Convert views into a side-by-side video with the same output resolution as the input
8230 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
8235 Change the frame rate by interpolating new video output frames from the source
8238 This filter is not designed to function correctly with interlaced media. If
8239 you wish to change the frame rate of interlaced media then you are required
8240 to deinterlace before this filter and re-interlace after this filter.
8242 A description of the accepted options follows.
8246 Specify the output frames per second. This option can also be specified
8247 as a value alone. The default is @code{50}.
8250 Specify the start of a range where the output frame will be created as a
8251 linear interpolation of two frames. The range is [@code{0}-@code{255}],
8252 the default is @code{15}.
8255 Specify the end of a range where the output frame will be created as a
8256 linear interpolation of two frames. The range is [@code{0}-@code{255}],
8257 the default is @code{240}.
8260 Specify the level at which a scene change is detected as a value between
8261 0 and 100 to indicate a new scene; a low value reflects a low
8262 probability for the current frame to introduce a new scene, while a higher
8263 value means the current frame is more likely to be one.
8264 The default is @code{7}.
8267 Specify flags influencing the filter process.
8269 Available value for @var{flags} is:
8272 @item scene_change_detect, scd
8273 Enable scene change detection using the value of the option @var{scene}.
8274 This flag is enabled by default.
8280 Select one frame every N-th frame.
8282 This filter accepts the following option:
8285 Select frame after every @code{step} frames.
8286 Allowed values are positive integers higher than 0. Default value is @code{1}.
8292 Apply a frei0r effect to the input video.
8294 To enable the compilation of this filter, you need to install the frei0r
8295 header and configure FFmpeg with @code{--enable-frei0r}.
8297 It accepts the following parameters:
8302 The name of the frei0r effect to load. If the environment variable
8303 @env{FREI0R_PATH} is defined, the frei0r effect is searched for in each of the
8304 directories specified by the colon-separated list in @env{FREIOR_PATH}.
8305 Otherwise, the standard frei0r paths are searched, in this order:
8306 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
8307 @file{/usr/lib/frei0r-1/}.
8310 A '|'-separated list of parameters to pass to the frei0r effect.
8314 A frei0r effect parameter can be a boolean (its value is either
8315 "y" or "n"), a double, a color (specified as
8316 @var{R}/@var{G}/@var{B}, where @var{R}, @var{G}, and @var{B} are floating point
8317 numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
8318 section in the ffmpeg-utils manual), a position (specified as @var{X}/@var{Y}, where
8319 @var{X} and @var{Y} are floating point numbers) and/or a string.
8321 The number and types of parameters depend on the loaded effect. If an
8322 effect parameter is not specified, the default value is set.
8324 @subsection Examples
8328 Apply the distort0r effect, setting the first two double parameters:
8330 frei0r=filter_name=distort0r:filter_params=0.5|0.01
8334 Apply the colordistance effect, taking a color as the first parameter:
8336 frei0r=colordistance:0.2/0.3/0.4
8337 frei0r=colordistance:violet
8338 frei0r=colordistance:0x112233
8342 Apply the perspective effect, specifying the top left and top right image
8345 frei0r=perspective:0.2/0.2|0.8/0.2
8349 For more information, see
8350 @url{http://frei0r.dyne.org}
8354 Apply fast and simple postprocessing. It is a faster version of @ref{spp}.
8356 It splits (I)DCT into horizontal/vertical passes. Unlike the simple post-
8357 processing filter, one of them is performed once per block, not per pixel.
8358 This allows for much higher speed.
8360 The filter accepts the following options:
8364 Set quality. This option defines the number of levels for averaging. It accepts
8365 an integer in the range 4-5. Default value is @code{4}.
8368 Force a constant quantization parameter. It accepts an integer in range 0-63.
8369 If not set, the filter will use the QP from the video stream (if available).
8372 Set filter strength. It accepts an integer in range -15 to 32. Lower values mean
8373 more details but also more artifacts, while higher values make the image smoother
8374 but also blurrier. Default value is @code{0} − PSNR optimal.
8377 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
8378 option may cause flicker since the B-Frames have often larger QP. Default is
8379 @code{0} (not enabled).
8385 Apply Gaussian blur filter.
8387 The filter accepts the following options:
8391 Set horizontal sigma, standard deviation of Gaussian blur. Default is @code{0.5}.
8394 Set number of steps for Gaussian approximation. Defauls is @code{1}.
8397 Set which planes to filter. By default all planes are filtered.
8400 Set vertical sigma, if negative it will be same as @code{sigma}.
8401 Default is @code{-1}.
8406 The filter accepts the following options:
8410 Set the luminance expression.
8412 Set the chrominance blue expression.
8414 Set the chrominance red expression.
8416 Set the alpha expression.
8418 Set the red expression.
8420 Set the green expression.
8422 Set the blue expression.
8425 The colorspace is selected according to the specified options. If one
8426 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
8427 options is specified, the filter will automatically select a YCbCr
8428 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
8429 @option{blue_expr} options is specified, it will select an RGB
8432 If one of the chrominance expression is not defined, it falls back on the other
8433 one. If no alpha expression is specified it will evaluate to opaque value.
8434 If none of chrominance expressions are specified, they will evaluate
8435 to the luminance expression.
8437 The expressions can use the following variables and functions:
8441 The sequential number of the filtered frame, starting from @code{0}.
8445 The coordinates of the current sample.
8449 The width and height of the image.
8453 Width and height scale depending on the currently filtered plane. It is the
8454 ratio between the corresponding luma plane number of pixels and the current
8455 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
8456 @code{0.5,0.5} for chroma planes.
8459 Time of the current frame, expressed in seconds.
8462 Return the value of the pixel at location (@var{x},@var{y}) of the current
8466 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
8470 Return the value of the pixel at location (@var{x},@var{y}) of the
8471 blue-difference chroma plane. Return 0 if there is no such plane.
8474 Return the value of the pixel at location (@var{x},@var{y}) of the
8475 red-difference chroma plane. Return 0 if there is no such plane.
8480 Return the value of the pixel at location (@var{x},@var{y}) of the
8481 red/green/blue component. Return 0 if there is no such component.
8484 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
8485 plane. Return 0 if there is no such plane.
8488 For functions, if @var{x} and @var{y} are outside the area, the value will be
8489 automatically clipped to the closer edge.
8491 @subsection Examples
8495 Flip the image horizontally:
8501 Generate a bidimensional sine wave, with angle @code{PI/3} and a
8502 wavelength of 100 pixels:
8504 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
8508 Generate a fancy enigmatic moving light:
8510 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
8514 Generate a quick emboss effect:
8516 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
8520 Modify RGB components depending on pixel position:
8522 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
8526 Create a radial gradient that is the same size as the input (also see
8527 the @ref{vignette} filter):
8529 geq=lum=255*gauss((X/W-0.5)*3)*gauss((Y/H-0.5)*3)/gauss(0)/gauss(0),format=gray
8535 Fix the banding artifacts that are sometimes introduced into nearly flat
8536 regions by truncation to 8-bit color depth.
8537 Interpolate the gradients that should go where the bands are, and
8540 It is designed for playback only. Do not use it prior to
8541 lossy compression, because compression tends to lose the dither and
8542 bring back the bands.
8544 It accepts the following parameters:
8549 The maximum amount by which the filter will change any one pixel. This is also
8550 the threshold for detecting nearly flat regions. Acceptable values range from
8551 .51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
8555 The neighborhood to fit the gradient to. A larger radius makes for smoother
8556 gradients, but also prevents the filter from modifying the pixels near detailed
8557 regions. Acceptable values are 8-32; the default value is 16. Out-of-range
8558 values will be clipped to the valid range.
8562 Alternatively, the options can be specified as a flat string:
8563 @var{strength}[:@var{radius}]
8565 @subsection Examples
8569 Apply the filter with a @code{3.5} strength and radius of @code{8}:
8575 Specify radius, omitting the strength (which will fall-back to the default
8586 Apply a Hald CLUT to a video stream.
8588 First input is the video stream to process, and second one is the Hald CLUT.
8589 The Hald CLUT input can be a simple picture or a complete video stream.
8591 The filter accepts the following options:
8595 Force termination when the shortest input terminates. Default is @code{0}.
8597 Continue applying the last CLUT after the end of the stream. A value of
8598 @code{0} disable the filter after the last frame of the CLUT is reached.
8599 Default is @code{1}.
8602 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
8603 filters share the same internals).
8605 More information about the Hald CLUT can be found on Eskil Steenberg's website
8606 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
8608 @subsection Workflow examples
8610 @subsubsection Hald CLUT video stream
8612 Generate an identity Hald CLUT stream altered with various effects:
8614 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
8617 Note: make sure you use a lossless codec.
8619 Then use it with @code{haldclut} to apply it on some random stream:
8621 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
8624 The Hald CLUT will be applied to the 10 first seconds (duration of
8625 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
8626 to the remaining frames of the @code{mandelbrot} stream.
8628 @subsubsection Hald CLUT with preview
8630 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
8631 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
8632 biggest possible square starting at the top left of the picture. The remaining
8633 padding pixels (bottom or right) will be ignored. This area can be used to add
8634 a preview of the Hald CLUT.
8636 Typically, the following generated Hald CLUT will be supported by the
8637 @code{haldclut} filter:
8640 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
8641 pad=iw+320 [padded_clut];
8642 smptebars=s=320x256, split [a][b];
8643 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
8644 [main][b] overlay=W-320" -frames:v 1 clut.png
8647 It contains the original and a preview of the effect of the CLUT: SMPTE color
8648 bars are displayed on the right-top, and below the same color bars processed by
8651 Then, the effect of this Hald CLUT can be visualized with:
8653 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
8658 Flip the input video horizontally.
8660 For example, to horizontally flip the input video with @command{ffmpeg}:
8662 ffmpeg -i in.avi -vf "hflip" out.avi
8666 This filter applies a global color histogram equalization on a
8669 It can be used to correct video that has a compressed range of pixel
8670 intensities. The filter redistributes the pixel intensities to
8671 equalize their distribution across the intensity range. It may be
8672 viewed as an "automatically adjusting contrast filter". This filter is
8673 useful only for correcting degraded or poorly captured source
8676 The filter accepts the following options:
8680 Determine the amount of equalization to be applied. As the strength
8681 is reduced, the distribution of pixel intensities more-and-more
8682 approaches that of the input frame. The value must be a float number
8683 in the range [0,1] and defaults to 0.200.
8686 Set the maximum intensity that can generated and scale the output
8687 values appropriately. The strength should be set as desired and then
8688 the intensity can be limited if needed to avoid washing-out. The value
8689 must be a float number in the range [0,1] and defaults to 0.210.
8692 Set the antibanding level. If enabled the filter will randomly vary
8693 the luminance of output pixels by a small amount to avoid banding of
8694 the histogram. Possible values are @code{none}, @code{weak} or
8695 @code{strong}. It defaults to @code{none}.
8700 Compute and draw a color distribution histogram for the input video.
8702 The computed histogram is a representation of the color component
8703 distribution in an image.
8705 Standard histogram displays the color components distribution in an image.
8706 Displays color graph for each color component. Shows distribution of
8707 the Y, U, V, A or R, G, B components, depending on input format, in the
8708 current frame. Below each graph a color component scale meter is shown.
8710 The filter accepts the following options:
8714 Set height of level. Default value is @code{200}.
8715 Allowed range is [50, 2048].
8718 Set height of color scale. Default value is @code{12}.
8719 Allowed range is [0, 40].
8723 It accepts the following values:
8726 Per color component graphs are placed below each other.
8729 Presents information identical to that in the @code{parade}, except
8730 that the graphs representing color components are superimposed directly
8733 Default is @code{parade}.
8736 Set mode. Can be either @code{linear}, or @code{logarithmic}.
8737 Default is @code{linear}.
8740 Set what color components to display.
8741 Default is @code{7}.
8744 Set foreground opacity. Default is @code{0.7}.
8747 Set background opacity. Default is @code{0.5}.
8750 @subsection Examples
8755 Calculate and draw histogram:
8757 ffplay -i input -vf histogram
8765 This is a high precision/quality 3d denoise filter. It aims to reduce
8766 image noise, producing smooth images and making still images really
8767 still. It should enhance compressibility.
8769 It accepts the following optional parameters:
8773 A non-negative floating point number which specifies spatial luma strength.
8776 @item chroma_spatial
8777 A non-negative floating point number which specifies spatial chroma strength.
8778 It defaults to 3.0*@var{luma_spatial}/4.0.
8781 A floating point number which specifies luma temporal strength. It defaults to
8782 6.0*@var{luma_spatial}/4.0.
8785 A floating point number which specifies chroma temporal strength. It defaults to
8786 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}.
8789 @anchor{hwupload_cuda}
8790 @section hwupload_cuda
8792 Upload system memory frames to a CUDA device.
8794 It accepts the following optional parameters:
8798 The number of the CUDA device to use
8803 Apply a high-quality magnification filter designed for pixel art. This filter
8804 was originally created by Maxim Stepin.
8806 It accepts the following option:
8810 Set the scaling dimension: @code{2} for @code{hq2x}, @code{3} for
8811 @code{hq3x} and @code{4} for @code{hq4x}.
8812 Default is @code{3}.
8816 Stack input videos horizontally.
8818 All streams must be of same pixel format and of same height.
8820 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
8821 to create same output.
8823 The filter accept the following option:
8827 Set number of input streams. Default is 2.
8830 If set to 1, force the output to terminate when the shortest input
8831 terminates. Default value is 0.
8836 Modify the hue and/or the saturation of the input.
8838 It accepts the following parameters:
8842 Specify the hue angle as a number of degrees. It accepts an expression,
8843 and defaults to "0".
8846 Specify the saturation in the [-10,10] range. It accepts an expression and
8850 Specify the hue angle as a number of radians. It accepts an
8851 expression, and defaults to "0".
8854 Specify the brightness in the [-10,10] range. It accepts an expression and
8858 @option{h} and @option{H} are mutually exclusive, and can't be
8859 specified at the same time.
8861 The @option{b}, @option{h}, @option{H} and @option{s} option values are
8862 expressions containing the following constants:
8866 frame count of the input frame starting from 0
8869 presentation timestamp of the input frame expressed in time base units
8872 frame rate of the input video, NAN if the input frame rate is unknown
8875 timestamp expressed in seconds, NAN if the input timestamp is unknown
8878 time base of the input video
8881 @subsection Examples
8885 Set the hue to 90 degrees and the saturation to 1.0:
8891 Same command but expressing the hue in radians:
8897 Rotate hue and make the saturation swing between 0
8898 and 2 over a period of 1 second:
8900 hue="H=2*PI*t: s=sin(2*PI*t)+1"
8904 Apply a 3 seconds saturation fade-in effect starting at 0:
8909 The general fade-in expression can be written as:
8911 hue="s=min(0\, max((t-START)/DURATION\, 1))"
8915 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
8917 hue="s=max(0\, min(1\, (8-t)/3))"
8920 The general fade-out expression can be written as:
8922 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
8927 @subsection Commands
8929 This filter supports the following commands:
8935 Modify the hue and/or the saturation and/or brightness of the input video.
8936 The command accepts the same syntax of the corresponding option.
8938 If the specified expression is not valid, it is kept at its current
8944 Grow first stream into second stream by connecting components.
8945 This makes it possible to build more robust edge masks.
8947 This filter accepts the following options:
8951 Set which planes will be processed as bitmap, unprocessed planes will be
8952 copied from first stream.
8953 By default value 0xf, all planes will be processed.
8956 Set threshold which is used in filtering. If pixel component value is higher than
8957 this value filter algorithm for connecting components is activated.
8958 By default value is 0.
8963 Detect video interlacing type.
8965 This filter tries to detect if the input frames are interlaced, progressive,
8966 top or bottom field first. It will also try to detect fields that are
8967 repeated between adjacent frames (a sign of telecine).
8969 Single frame detection considers only immediately adjacent frames when classifying each frame.
8970 Multiple frame detection incorporates the classification history of previous frames.
8972 The filter will log these metadata values:
8975 @item single.current_frame
8976 Detected type of current frame using single-frame detection. One of:
8977 ``tff'' (top field first), ``bff'' (bottom field first),
8978 ``progressive'', or ``undetermined''
8981 Cumulative number of frames detected as top field first using single-frame detection.
8984 Cumulative number of frames detected as top field first using multiple-frame detection.
8987 Cumulative number of frames detected as bottom field first using single-frame detection.
8989 @item multiple.current_frame
8990 Detected type of current frame using multiple-frame detection. One of:
8991 ``tff'' (top field first), ``bff'' (bottom field first),
8992 ``progressive'', or ``undetermined''
8995 Cumulative number of frames detected as bottom field first using multiple-frame detection.
8997 @item single.progressive
8998 Cumulative number of frames detected as progressive using single-frame detection.
9000 @item multiple.progressive
9001 Cumulative number of frames detected as progressive using multiple-frame detection.
9003 @item single.undetermined
9004 Cumulative number of frames that could not be classified using single-frame detection.
9006 @item multiple.undetermined
9007 Cumulative number of frames that could not be classified using multiple-frame detection.
9009 @item repeated.current_frame
9010 Which field in the current frame is repeated from the last. One of ``neither'', ``top'', or ``bottom''.
9012 @item repeated.neither
9013 Cumulative number of frames with no repeated field.
9016 Cumulative number of frames with the top field repeated from the previous frame's top field.
9018 @item repeated.bottom
9019 Cumulative number of frames with the bottom field repeated from the previous frame's bottom field.
9022 The filter accepts the following options:
9026 Set interlacing threshold.
9028 Set progressive threshold.
9030 Threshold for repeated field detection.
9032 Number of frames after which a given frame's contribution to the
9033 statistics is halved (i.e., it contributes only 0.5 to its
9034 classification). The default of 0 means that all frames seen are given
9035 full weight of 1.0 forever.
9036 @item analyze_interlaced_flag
9037 When this is not 0 then idet will use the specified number of frames to determine
9038 if the interlaced flag is accurate, it will not count undetermined frames.
9039 If the flag is found to be accurate it will be used without any further
9040 computations, if it is found to be inaccurate it will be cleared without any
9041 further computations. This allows inserting the idet filter as a low computational
9042 method to clean up the interlaced flag
9047 Deinterleave or interleave fields.
9049 This filter allows one to process interlaced images fields without
9050 deinterlacing them. Deinterleaving splits the input frame into 2
9051 fields (so called half pictures). Odd lines are moved to the top
9052 half of the output image, even lines to the bottom half.
9053 You can process (filter) them independently and then re-interleave them.
9055 The filter accepts the following options:
9059 @item chroma_mode, c
9061 Available values for @var{luma_mode}, @var{chroma_mode} and
9062 @var{alpha_mode} are:
9068 @item deinterleave, d
9069 Deinterleave fields, placing one above the other.
9072 Interleave fields. Reverse the effect of deinterleaving.
9074 Default value is @code{none}.
9077 @item chroma_swap, cs
9078 @item alpha_swap, as
9079 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
9084 Apply inflate effect to the video.
9086 This filter replaces the pixel by the local(3x3) average by taking into account
9087 only values higher than the pixel.
9089 It accepts the following options:
9096 Limit the maximum change for each plane, default is 65535.
9097 If 0, plane will remain unchanged.
9102 Simple interlacing filter from progressive contents. This interleaves upper (or
9103 lower) lines from odd frames with lower (or upper) lines from even frames,
9104 halving the frame rate and preserving image height.
9107 Original Original New Frame
9108 Frame 'j' Frame 'j+1' (tff)
9109 ========== =========== ==================
9110 Line 0 --------------------> Frame 'j' Line 0
9111 Line 1 Line 1 ----> Frame 'j+1' Line 1
9112 Line 2 ---------------------> Frame 'j' Line 2
9113 Line 3 Line 3 ----> Frame 'j+1' Line 3
9115 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
9118 It accepts the following optional parameters:
9122 This determines whether the interlaced frame is taken from the even
9123 (tff - default) or odd (bff) lines of the progressive frame.
9126 Enable (default) or disable the vertical lowpass filter to avoid twitter
9127 interlacing and reduce moire patterns.
9132 Deinterlace input video by applying Donald Graft's adaptive kernel
9133 deinterling. Work on interlaced parts of a video to produce
9136 The description of the accepted parameters follows.
9140 Set the threshold which affects the filter's tolerance when
9141 determining if a pixel line must be processed. It must be an integer
9142 in the range [0,255] and defaults to 10. A value of 0 will result in
9143 applying the process on every pixels.
9146 Paint pixels exceeding the threshold value to white if set to 1.
9150 Set the fields order. Swap fields if set to 1, leave fields alone if
9154 Enable additional sharpening if set to 1. Default is 0.
9157 Enable twoway sharpening if set to 1. Default is 0.
9160 @subsection Examples
9164 Apply default values:
9166 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
9170 Enable additional sharpening:
9176 Paint processed pixels in white:
9182 @section lenscorrection
9184 Correct radial lens distortion
9186 This filter can be used to correct for radial distortion as can result from the use
9187 of wide angle lenses, and thereby re-rectify the image. To find the right parameters
9188 one can use tools available for example as part of opencv or simply trial-and-error.
9189 To use opencv use the calibration sample (under samples/cpp) from the opencv sources
9190 and extract the k1 and k2 coefficients from the resulting matrix.
9192 Note that effectively the same filter is available in the open-source tools Krita and
9193 Digikam from the KDE project.
9195 In contrast to the @ref{vignette} filter, which can also be used to compensate lens errors,
9196 this filter corrects the distortion of the image, whereas @ref{vignette} corrects the
9197 brightness distribution, so you may want to use both filters together in certain
9198 cases, though you will have to take care of ordering, i.e. whether vignetting should
9199 be applied before or after lens correction.
9203 The filter accepts the following options:
9207 Relative x-coordinate of the focal point of the image, and thereby the center of the
9208 distortion. This value has a range [0,1] and is expressed as fractions of the image
9211 Relative y-coordinate of the focal point of the image, and thereby the center of the
9212 distortion. This value has a range [0,1] and is expressed as fractions of the image
9215 Coefficient of the quadratic correction term. 0.5 means no correction.
9217 Coefficient of the double quadratic correction term. 0.5 means no correction.
9220 The formula that generates the correction is:
9222 @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)
9224 where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the
9225 distances from the focal point in the source and target images, respectively.
9231 The filter accepts the following options:
9235 Set the number of loops.
9238 Set maximal size in number of frames.
9241 Set first frame of loop.
9247 Apply a 3D LUT to an input video.
9249 The filter accepts the following options:
9253 Set the 3D LUT file name.
9255 Currently supported formats:
9267 Select interpolation mode.
9269 Available values are:
9273 Use values from the nearest defined point.
9275 Interpolate values using the 8 points defining a cube.
9277 Interpolate values using a tetrahedron.
9281 @section lut, lutrgb, lutyuv
9283 Compute a look-up table for binding each pixel component input value
9284 to an output value, and apply it to the input video.
9286 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
9287 to an RGB input video.
9289 These filters accept the following parameters:
9292 set first pixel component expression
9294 set second pixel component expression
9296 set third pixel component expression
9298 set fourth pixel component expression, corresponds to the alpha component
9301 set red component expression
9303 set green component expression
9305 set blue component expression
9307 alpha component expression
9310 set Y/luminance component expression
9312 set U/Cb component expression
9314 set V/Cr component expression
9317 Each of them specifies the expression to use for computing the lookup table for
9318 the corresponding pixel component values.
9320 The exact component associated to each of the @var{c*} options depends on the
9323 The @var{lut} filter requires either YUV or RGB pixel formats in input,
9324 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
9326 The expressions can contain the following constants and functions:
9331 The input width and height.
9334 The input value for the pixel component.
9337 The input value, clipped to the @var{minval}-@var{maxval} range.
9340 The maximum value for the pixel component.
9343 The minimum value for the pixel component.
9346 The negated value for the pixel component value, clipped to the
9347 @var{minval}-@var{maxval} range; it corresponds to the expression
9348 "maxval-clipval+minval".
9351 The computed value in @var{val}, clipped to the
9352 @var{minval}-@var{maxval} range.
9354 @item gammaval(gamma)
9355 The computed gamma correction value of the pixel component value,
9356 clipped to the @var{minval}-@var{maxval} range. It corresponds to the
9358 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
9362 All expressions default to "val".
9364 @subsection Examples
9370 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
9371 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
9374 The above is the same as:
9376 lutrgb="r=negval:g=negval:b=negval"
9377 lutyuv="y=negval:u=negval:v=negval"
9387 Remove chroma components, turning the video into a graytone image:
9389 lutyuv="u=128:v=128"
9393 Apply a luma burning effect:
9399 Remove green and blue components:
9405 Set a constant alpha channel value on input:
9407 format=rgba,lutrgb=a="maxval-minval/2"
9411 Correct luminance gamma by a factor of 0.5:
9413 lutyuv=y=gammaval(0.5)
9417 Discard least significant bits of luma:
9419 lutyuv=y='bitand(val, 128+64+32)'
9423 Technicolor like effect:
9425 lutyuv=u='(val-maxval/2)*2+maxval/2':v='(val-maxval/2)*2+maxval/2'
9431 Compute and apply a lookup table from two video inputs.
9433 This filter accepts the following parameters:
9436 set first pixel component expression
9438 set second pixel component expression
9440 set third pixel component expression
9442 set fourth pixel component expression, corresponds to the alpha component
9445 Each of them specifies the expression to use for computing the lookup table for
9446 the corresponding pixel component values.
9448 The exact component associated to each of the @var{c*} options depends on the
9451 The expressions can contain the following constants:
9456 The input width and height.
9459 The first input value for the pixel component.
9462 The second input value for the pixel component.
9465 The first input video bit depth.
9468 The second input video bit depth.
9471 All expressions default to "x".
9473 @subsection Examples
9477 Highlight differences between two RGB video streams:
9479 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)'
9483 Highlight differences between two YUV video streams:
9485 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)'
9489 @section maskedclamp
9491 Clamp the first input stream with the second input and third input stream.
9493 Returns the value of first stream to be between second input
9494 stream - @code{undershoot} and third input stream + @code{overshoot}.
9496 This filter accepts the following options:
9499 Default value is @code{0}.
9502 Default value is @code{0}.
9505 Set which planes will be processed as bitmap, unprocessed planes will be
9506 copied from first stream.
9507 By default value 0xf, all planes will be processed.
9510 @section maskedmerge
9512 Merge the first input stream with the second input stream using per pixel
9513 weights in the third input stream.
9515 A value of 0 in the third stream pixel component means that pixel component
9516 from first stream is returned unchanged, while maximum value (eg. 255 for
9517 8-bit videos) means that pixel component from second stream is returned
9518 unchanged. Intermediate values define the amount of merging between both
9519 input stream's pixel components.
9521 This filter accepts the following options:
9524 Set which planes will be processed as bitmap, unprocessed planes will be
9525 copied from first stream.
9526 By default value 0xf, all planes will be processed.
9531 Apply motion-compensation deinterlacing.
9533 It needs one field per frame as input and must thus be used together
9534 with yadif=1/3 or equivalent.
9536 This filter accepts the following options:
9539 Set the deinterlacing mode.
9541 It accepts one of the following values:
9546 use iterative motion estimation
9548 like @samp{slow}, but use multiple reference frames.
9550 Default value is @samp{fast}.
9553 Set the picture field parity assumed for the input video. It must be
9554 one of the following values:
9558 assume top field first
9560 assume bottom field first
9563 Default value is @samp{bff}.
9566 Set per-block quantization parameter (QP) used by the internal
9569 Higher values should result in a smoother motion vector field but less
9570 optimal individual vectors. Default value is 1.
9573 @section mergeplanes
9575 Merge color channel components from several video streams.
9577 The filter accepts up to 4 input streams, and merge selected input
9578 planes to the output video.
9580 This filter accepts the following options:
9583 Set input to output plane mapping. Default is @code{0}.
9585 The mappings is specified as a bitmap. It should be specified as a
9586 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
9587 mapping for the first plane of the output stream. 'A' sets the number of
9588 the input stream to use (from 0 to 3), and 'a' the plane number of the
9589 corresponding input to use (from 0 to 3). The rest of the mappings is
9590 similar, 'Bb' describes the mapping for the output stream second
9591 plane, 'Cc' describes the mapping for the output stream third plane and
9592 'Dd' describes the mapping for the output stream fourth plane.
9595 Set output pixel format. Default is @code{yuva444p}.
9598 @subsection Examples
9602 Merge three gray video streams of same width and height into single video stream:
9604 [a0][a1][a2]mergeplanes=0x001020:yuv444p
9608 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
9610 [a0][a1]mergeplanes=0x00010210:yuva444p
9614 Swap Y and A plane in yuva444p stream:
9616 format=yuva444p,mergeplanes=0x03010200:yuva444p
9620 Swap U and V plane in yuv420p stream:
9622 format=yuv420p,mergeplanes=0x000201:yuv420p
9626 Cast a rgb24 clip to yuv444p:
9628 format=rgb24,mergeplanes=0x000102:yuv444p
9634 Estimate and export motion vectors using block matching algorithms.
9635 Motion vectors are stored in frame side data to be used by other filters.
9637 This filter accepts the following options:
9640 Specify the motion estimation method. Accepts one of the following values:
9644 Exhaustive search algorithm.
9646 Three step search algorithm.
9648 Two dimensional logarithmic search algorithm.
9650 New three step search algorithm.
9652 Four step search algorithm.
9654 Diamond search algorithm.
9656 Hexagon-based search algorithm.
9658 Enhanced predictive zonal search algorithm.
9660 Uneven multi-hexagon search algorithm.
9662 Default value is @samp{esa}.
9665 Macroblock size. Default @code{16}.
9668 Search parameter. Default @code{7}.
9671 @section midequalizer
9673 Apply Midway Image Equalization effect using two video streams.
9675 Midway Image Equalization adjusts a pair of images to have the same
9676 histogram, while maintaining their dynamics as much as possible. It's
9677 useful for e.g. matching exposures from a pair of stereo cameras.
9679 This filter has two inputs and one output, which must be of same pixel format, but
9680 may be of different sizes. The output of filter is first input adjusted with
9681 midway histogram of both inputs.
9683 This filter accepts the following option:
9687 Set which planes to process. Default is @code{15}, which is all available planes.
9690 @section minterpolate
9692 Convert the video to specified frame rate using motion interpolation.
9694 This filter accepts the following options:
9697 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}.
9700 Motion interpolation mode. Following values are accepted:
9703 Duplicate previous or next frame for interpolating new ones.
9705 Blend source frames. Interpolated frame is mean of previous and next frames.
9707 Motion compensated interpolation. Following options are effective when this mode is selected:
9711 Motion compensation mode. Following values are accepted:
9714 Overlapped block motion compensation.
9716 Adaptive overlapped block motion compensation. Window weighting coefficients are controlled adaptively according to the reliabilities of the neighboring motion vectors to reduce oversmoothing.
9718 Default mode is @samp{obmc}.
9721 Motion estimation mode. Following values are accepted:
9724 Bidirectional motion estimation. Motion vectors are estimated for each source frame in both forward and backward directions.
9726 Bilateral motion estimation. Motion vectors are estimated directly for interpolated frame.
9728 Default mode is @samp{bilat}.
9731 The algorithm to be used for motion estimation. Following values are accepted:
9734 Exhaustive search algorithm.
9736 Three step search algorithm.
9738 Two dimensional logarithmic search algorithm.
9740 New three step search algorithm.
9742 Four step search algorithm.
9744 Diamond search algorithm.
9746 Hexagon-based search algorithm.
9748 Enhanced predictive zonal search algorithm.
9750 Uneven multi-hexagon search algorithm.
9752 Default algorithm is @samp{epzs}.
9755 Macroblock size. Default @code{16}.
9758 Motion estimation search parameter. Default @code{32}.
9761 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).
9766 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:
9769 Disable scene change detection.
9771 Frame difference. Corresponding pixel values are compared and if it satisfies @var{scd_threshold} scene change is detected.
9773 Default method is @samp{fdiff}.
9776 Scene change detection threshold. Default is @code{5.0}.
9781 Drop frames that do not differ greatly from the previous frame in
9782 order to reduce frame rate.
9784 The main use of this filter is for very-low-bitrate encoding
9785 (e.g. streaming over dialup modem), but it could in theory be used for
9786 fixing movies that were inverse-telecined incorrectly.
9788 A description of the accepted options follows.
9792 Set the maximum number of consecutive frames which can be dropped (if
9793 positive), or the minimum interval between dropped frames (if
9794 negative). If the value is 0, the frame is dropped unregarding the
9795 number of previous sequentially dropped frames.
9802 Set the dropping threshold values.
9804 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
9805 represent actual pixel value differences, so a threshold of 64
9806 corresponds to 1 unit of difference for each pixel, or the same spread
9807 out differently over the block.
9809 A frame is a candidate for dropping if no 8x8 blocks differ by more
9810 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
9811 meaning the whole image) differ by more than a threshold of @option{lo}.
9813 Default value for @option{hi} is 64*12, default value for @option{lo} is
9814 64*5, and default value for @option{frac} is 0.33.
9822 It accepts an integer in input; if non-zero it negates the
9823 alpha component (if available). The default value in input is 0.
9827 Denoise frames using Non-Local Means algorithm.
9829 Each pixel is adjusted by looking for other pixels with similar contexts. This
9830 context similarity is defined by comparing their surrounding patches of size
9831 @option{p}x@option{p}. Patches are searched in an area of @option{r}x@option{r}
9834 Note that the research area defines centers for patches, which means some
9835 patches will be made of pixels outside that research area.
9837 The filter accepts the following options.
9841 Set denoising strength.
9847 Same as @option{p} but for chroma planes.
9849 The default value is @var{0} and means automatic.
9855 Same as @option{r} but for chroma planes.
9857 The default value is @var{0} and means automatic.
9862 Deinterlace video using neural network edge directed interpolation.
9864 This filter accepts the following options:
9868 Mandatory option, without binary file filter can not work.
9869 Currently file can be found here:
9870 https://github.com/dubhater/vapoursynth-nnedi3/blob/master/src/nnedi3_weights.bin
9873 Set which frames to deinterlace, by default it is @code{all}.
9874 Can be @code{all} or @code{interlaced}.
9877 Set mode of operation.
9879 Can be one of the following:
9883 Use frame flags, both fields.
9885 Use frame flags, single field.
9889 Use bottom field only.
9891 Use both fields, top first.
9893 Use both fields, bottom first.
9897 Set which planes to process, by default filter process all frames.
9900 Set size of local neighborhood around each pixel, used by the predictor neural
9903 Can be one of the following:
9916 Set the number of neurons in predicctor neural network.
9917 Can be one of the following:
9928 Controls the number of different neural network predictions that are blended
9929 together to compute the final output value. Can be @code{fast}, default or
9933 Set which set of weights to use in the predictor.
9934 Can be one of the following:
9938 weights trained to minimize absolute error
9940 weights trained to minimize squared error
9944 Controls whether or not the prescreener neural network is used to decide
9945 which pixels should be processed by the predictor neural network and which
9946 can be handled by simple cubic interpolation.
9947 The prescreener is trained to know whether cubic interpolation will be
9948 sufficient for a pixel or whether it should be predicted by the predictor nn.
9949 The computational complexity of the prescreener nn is much less than that of
9950 the predictor nn. Since most pixels can be handled by cubic interpolation,
9951 using the prescreener generally results in much faster processing.
9952 The prescreener is pretty accurate, so the difference between using it and not
9953 using it is almost always unnoticeable.
9955 Can be one of the following:
9963 Default is @code{new}.
9966 Set various debugging flags.
9971 Force libavfilter not to use any of the specified pixel formats for the
9972 input to the next filter.
9974 It accepts the following parameters:
9978 A '|'-separated list of pixel format names, such as
9979 apix_fmts=yuv420p|monow|rgb24".
9983 @subsection Examples
9987 Force libavfilter to use a format different from @var{yuv420p} for the
9988 input to the vflip filter:
9990 noformat=pix_fmts=yuv420p,vflip
9994 Convert the input video to any of the formats not contained in the list:
9996 noformat=yuv420p|yuv444p|yuv410p
10002 Add noise on video input frame.
10004 The filter accepts the following options:
10012 Set noise seed for specific pixel component or all pixel components in case
10013 of @var{all_seed}. Default value is @code{123457}.
10015 @item all_strength, alls
10016 @item c0_strength, c0s
10017 @item c1_strength, c1s
10018 @item c2_strength, c2s
10019 @item c3_strength, c3s
10020 Set noise strength for specific pixel component or all pixel components in case
10021 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
10023 @item all_flags, allf
10024 @item c0_flags, c0f
10025 @item c1_flags, c1f
10026 @item c2_flags, c2f
10027 @item c3_flags, c3f
10028 Set pixel component flags or set flags for all components if @var{all_flags}.
10029 Available values for component flags are:
10032 averaged temporal noise (smoother)
10034 mix random noise with a (semi)regular pattern
10036 temporal noise (noise pattern changes between frames)
10038 uniform noise (gaussian otherwise)
10042 @subsection Examples
10044 Add temporal and uniform noise to input video:
10046 noise=alls=20:allf=t+u
10051 Pass the video source unchanged to the output.
10054 Optical Character Recognition
10056 This filter uses Tesseract for optical character recognition.
10058 It accepts the following options:
10062 Set datapath to tesseract data. Default is to use whatever was
10063 set at installation.
10066 Set language, default is "eng".
10069 Set character whitelist.
10072 Set character blacklist.
10075 The filter exports recognized text as the frame metadata @code{lavfi.ocr.text}.
10079 Apply a video transform using libopencv.
10081 To enable this filter, install the libopencv library and headers and
10082 configure FFmpeg with @code{--enable-libopencv}.
10084 It accepts the following parameters:
10089 The name of the libopencv filter to apply.
10091 @item filter_params
10092 The parameters to pass to the libopencv filter. If not specified, the default
10093 values are assumed.
10097 Refer to the official libopencv documentation for more precise
10099 @url{http://docs.opencv.org/master/modules/imgproc/doc/filtering.html}
10101 Several libopencv filters are supported; see the following subsections.
10106 Dilate an image by using a specific structuring element.
10107 It corresponds to the libopencv function @code{cvDilate}.
10109 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
10111 @var{struct_el} represents a structuring element, and has the syntax:
10112 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
10114 @var{cols} and @var{rows} represent the number of columns and rows of
10115 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
10116 point, and @var{shape} the shape for the structuring element. @var{shape}
10117 must be "rect", "cross", "ellipse", or "custom".
10119 If the value for @var{shape} is "custom", it must be followed by a
10120 string of the form "=@var{filename}". The file with name
10121 @var{filename} is assumed to represent a binary image, with each
10122 printable character corresponding to a bright pixel. When a custom
10123 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
10124 or columns and rows of the read file are assumed instead.
10126 The default value for @var{struct_el} is "3x3+0x0/rect".
10128 @var{nb_iterations} specifies the number of times the transform is
10129 applied to the image, and defaults to 1.
10133 # Use the default values
10136 # Dilate using a structuring element with a 5x5 cross, iterating two times
10137 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
10139 # Read the shape from the file diamond.shape, iterating two times.
10140 # The file diamond.shape may contain a pattern of characters like this
10146 # The specified columns and rows are ignored
10147 # but the anchor point coordinates are not
10148 ocv=dilate:0x0+2x2/custom=diamond.shape|2
10153 Erode an image by using a specific structuring element.
10154 It corresponds to the libopencv function @code{cvErode}.
10156 It accepts the parameters: @var{struct_el}:@var{nb_iterations},
10157 with the same syntax and semantics as the @ref{dilate} filter.
10161 Smooth the input video.
10163 The filter takes the following parameters:
10164 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
10166 @var{type} is the type of smooth filter to apply, and must be one of
10167 the following values: "blur", "blur_no_scale", "median", "gaussian",
10168 or "bilateral". The default value is "gaussian".
10170 The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
10171 depend on the smooth type. @var{param1} and
10172 @var{param2} accept integer positive values or 0. @var{param3} and
10173 @var{param4} accept floating point values.
10175 The default value for @var{param1} is 3. The default value for the
10176 other parameters is 0.
10178 These parameters correspond to the parameters assigned to the
10179 libopencv function @code{cvSmooth}.
10184 Overlay one video on top of another.
10186 It takes two inputs and has one output. The first input is the "main"
10187 video on which the second input is overlaid.
10189 It accepts the following parameters:
10191 A description of the accepted options follows.
10196 Set the expression for the x and y coordinates of the overlaid video
10197 on the main video. Default value is "0" for both expressions. In case
10198 the expression is invalid, it is set to a huge value (meaning that the
10199 overlay will not be displayed within the output visible area).
10202 The action to take when EOF is encountered on the secondary input; it accepts
10203 one of the following values:
10207 Repeat the last frame (the default).
10211 Pass the main input through.
10215 Set when the expressions for @option{x}, and @option{y} are evaluated.
10217 It accepts the following values:
10220 only evaluate expressions once during the filter initialization or
10221 when a command is processed
10224 evaluate expressions for each incoming frame
10227 Default value is @samp{frame}.
10230 If set to 1, force the output to terminate when the shortest input
10231 terminates. Default value is 0.
10234 Set the format for the output video.
10236 It accepts the following values:
10239 force YUV420 output
10242 force YUV422 output
10245 force YUV444 output
10248 force packed RGB output
10251 force planar RGB output
10254 Default value is @samp{yuv420}.
10256 @item rgb @emph{(deprecated)}
10257 If set to 1, force the filter to accept inputs in the RGB
10258 color space. Default value is 0. This option is deprecated, use
10259 @option{format} instead.
10262 If set to 1, force the filter to draw the last overlay frame over the
10263 main input until the end of the stream. A value of 0 disables this
10264 behavior. Default value is 1.
10267 The @option{x}, and @option{y} expressions can contain the following
10273 The main input width and height.
10277 The overlay input width and height.
10281 The computed values for @var{x} and @var{y}. They are evaluated for
10286 horizontal and vertical chroma subsample values of the output
10287 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
10291 the number of input frame, starting from 0
10294 the position in the file of the input frame, NAN if unknown
10297 The timestamp, expressed in seconds. It's NAN if the input timestamp is unknown.
10301 Note that the @var{n}, @var{pos}, @var{t} variables are available only
10302 when evaluation is done @emph{per frame}, and will evaluate to NAN
10303 when @option{eval} is set to @samp{init}.
10305 Be aware that frames are taken from each input video in timestamp
10306 order, hence, if their initial timestamps differ, it is a good idea
10307 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
10308 have them begin in the same zero timestamp, as the example for
10309 the @var{movie} filter does.
10311 You can chain together more overlays but you should test the
10312 efficiency of such approach.
10314 @subsection Commands
10316 This filter supports the following commands:
10320 Modify the x and y of the overlay input.
10321 The command accepts the same syntax of the corresponding option.
10323 If the specified expression is not valid, it is kept at its current
10327 @subsection Examples
10331 Draw the overlay at 10 pixels from the bottom right corner of the main
10334 overlay=main_w-overlay_w-10:main_h-overlay_h-10
10337 Using named options the example above becomes:
10339 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
10343 Insert a transparent PNG logo in the bottom left corner of the input,
10344 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
10346 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
10350 Insert 2 different transparent PNG logos (second logo on bottom
10351 right corner) using the @command{ffmpeg} tool:
10353 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
10357 Add a transparent color layer on top of the main video; @code{WxH}
10358 must specify the size of the main input to the overlay filter:
10360 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
10364 Play an original video and a filtered version (here with the deshake
10365 filter) side by side using the @command{ffplay} tool:
10367 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
10370 The above command is the same as:
10372 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
10376 Make a sliding overlay appearing from the left to the right top part of the
10377 screen starting since time 2:
10379 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
10383 Compose output by putting two input videos side to side:
10385 ffmpeg -i left.avi -i right.avi -filter_complex "
10386 nullsrc=size=200x100 [background];
10387 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
10388 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
10389 [background][left] overlay=shortest=1 [background+left];
10390 [background+left][right] overlay=shortest=1:x=100 [left+right]
10395 Mask 10-20 seconds of a video by applying the delogo filter to a section
10397 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
10398 -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]'
10403 Chain several overlays in cascade:
10405 nullsrc=s=200x200 [bg];
10406 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
10407 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
10408 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
10409 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
10410 [in3] null, [mid2] overlay=100:100 [out0]
10417 Apply Overcomplete Wavelet denoiser.
10419 The filter accepts the following options:
10425 Larger depth values will denoise lower frequency components more, but
10426 slow down filtering.
10428 Must be an int in the range 8-16, default is @code{8}.
10430 @item luma_strength, ls
10433 Must be a double value in the range 0-1000, default is @code{1.0}.
10435 @item chroma_strength, cs
10436 Set chroma strength.
10438 Must be a double value in the range 0-1000, default is @code{1.0}.
10444 Add paddings to the input image, and place the original input at the
10445 provided @var{x}, @var{y} coordinates.
10447 It accepts the following parameters:
10452 Specify an expression for the size of the output image with the
10453 paddings added. If the value for @var{width} or @var{height} is 0, the
10454 corresponding input size is used for the output.
10456 The @var{width} expression can reference the value set by the
10457 @var{height} expression, and vice versa.
10459 The default value of @var{width} and @var{height} is 0.
10463 Specify the offsets to place the input image at within the padded area,
10464 with respect to the top/left border of the output image.
10466 The @var{x} expression can reference the value set by the @var{y}
10467 expression, and vice versa.
10469 The default value of @var{x} and @var{y} is 0.
10471 If @var{x} or @var{y} evaluate to a negative number, they'll be changed
10472 so the input image is centered on the padded area.
10475 Specify the color of the padded area. For the syntax of this option,
10476 check the "Color" section in the ffmpeg-utils manual.
10478 The default value of @var{color} is "black".
10481 Specify when to evaluate @var{width}, @var{height}, @var{x} and @var{y} expression.
10483 It accepts the following values:
10487 Only evaluate expressions once during the filter initialization or when
10488 a command is processed.
10491 Evaluate expressions for each incoming frame.
10495 Default value is @samp{init}.
10498 Pad to aspect instead to a resolution.
10502 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
10503 options are expressions containing the following constants:
10508 The input video width and height.
10512 These are the same as @var{in_w} and @var{in_h}.
10516 The output width and height (the size of the padded area), as
10517 specified by the @var{width} and @var{height} expressions.
10521 These are the same as @var{out_w} and @var{out_h}.
10525 The x and y offsets as specified by the @var{x} and @var{y}
10526 expressions, or NAN if not yet specified.
10529 same as @var{iw} / @var{ih}
10532 input sample aspect ratio
10535 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
10539 The horizontal and vertical chroma subsample values. For example for the
10540 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
10543 @subsection Examples
10547 Add paddings with the color "violet" to the input video. The output video
10548 size is 640x480, and the top-left corner of the input video is placed at
10551 pad=640:480:0:40:violet
10554 The example above is equivalent to the following command:
10556 pad=width=640:height=480:x=0:y=40:color=violet
10560 Pad the input to get an output with dimensions increased by 3/2,
10561 and put the input video at the center of the padded area:
10563 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
10567 Pad the input to get a squared output with size equal to the maximum
10568 value between the input width and height, and put the input video at
10569 the center of the padded area:
10571 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
10575 Pad the input to get a final w/h ratio of 16:9:
10577 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
10581 In case of anamorphic video, in order to set the output display aspect
10582 correctly, it is necessary to use @var{sar} in the expression,
10583 according to the relation:
10585 (ih * X / ih) * sar = output_dar
10586 X = output_dar / sar
10589 Thus the previous example needs to be modified to:
10591 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
10595 Double the output size and put the input video in the bottom-right
10596 corner of the output padded area:
10598 pad="2*iw:2*ih:ow-iw:oh-ih"
10602 @anchor{palettegen}
10603 @section palettegen
10605 Generate one palette for a whole video stream.
10607 It accepts the following options:
10611 Set the maximum number of colors to quantize in the palette.
10612 Note: the palette will still contain 256 colors; the unused palette entries
10615 @item reserve_transparent
10616 Create a palette of 255 colors maximum and reserve the last one for
10617 transparency. Reserving the transparency color is useful for GIF optimization.
10618 If not set, the maximum of colors in the palette will be 256. You probably want
10619 to disable this option for a standalone image.
10623 Set statistics mode.
10625 It accepts the following values:
10628 Compute full frame histograms.
10630 Compute histograms only for the part that differs from previous frame. This
10631 might be relevant to give more importance to the moving part of your input if
10632 the background is static.
10634 Compute new histogram for each frame.
10637 Default value is @var{full}.
10640 The filter also exports the frame metadata @code{lavfi.color_quant_ratio}
10641 (@code{nb_color_in / nb_color_out}) which you can use to evaluate the degree of
10642 color quantization of the palette. This information is also visible at
10643 @var{info} logging level.
10645 @subsection Examples
10649 Generate a representative palette of a given video using @command{ffmpeg}:
10651 ffmpeg -i input.mkv -vf palettegen palette.png
10655 @section paletteuse
10657 Use a palette to downsample an input video stream.
10659 The filter takes two inputs: one video stream and a palette. The palette must
10660 be a 256 pixels image.
10662 It accepts the following options:
10666 Select dithering mode. Available algorithms are:
10669 Ordered 8x8 bayer dithering (deterministic)
10671 Dithering as defined by Paul Heckbert in 1982 (simple error diffusion).
10672 Note: this dithering is sometimes considered "wrong" and is included as a
10674 @item floyd_steinberg
10675 Floyd and Steingberg dithering (error diffusion)
10677 Frankie Sierra dithering v2 (error diffusion)
10679 Frankie Sierra dithering v2 "Lite" (error diffusion)
10682 Default is @var{sierra2_4a}.
10685 When @var{bayer} dithering is selected, this option defines the scale of the
10686 pattern (how much the crosshatch pattern is visible). A low value means more
10687 visible pattern for less banding, and higher value means less visible pattern
10688 at the cost of more banding.
10690 The option must be an integer value in the range [0,5]. Default is @var{2}.
10693 If set, define the zone to process
10697 Only the changing rectangle will be reprocessed. This is similar to GIF
10698 cropping/offsetting compression mechanism. This option can be useful for speed
10699 if only a part of the image is changing, and has use cases such as limiting the
10700 scope of the error diffusal @option{dither} to the rectangle that bounds the
10701 moving scene (it leads to more deterministic output if the scene doesn't change
10702 much, and as a result less moving noise and better GIF compression).
10705 Default is @var{none}.
10708 Take new palette for each output frame.
10711 @subsection Examples
10715 Use a palette (generated for example with @ref{palettegen}) to encode a GIF
10716 using @command{ffmpeg}:
10718 ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
10722 @section perspective
10724 Correct perspective of video not recorded perpendicular to the screen.
10726 A description of the accepted parameters follows.
10737 Set coordinates expression for top left, top right, bottom left and bottom right corners.
10738 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
10739 If the @code{sense} option is set to @code{source}, then the specified points will be sent
10740 to the corners of the destination. If the @code{sense} option is set to @code{destination},
10741 then the corners of the source will be sent to the specified coordinates.
10743 The expressions can use the following variables:
10748 the width and height of video frame.
10752 Output frame count.
10755 @item interpolation
10756 Set interpolation for perspective correction.
10758 It accepts the following values:
10764 Default value is @samp{linear}.
10767 Set interpretation of coordinate options.
10769 It accepts the following values:
10773 Send point in the source specified by the given coordinates to
10774 the corners of the destination.
10776 @item 1, destination
10778 Send the corners of the source to the point in the destination specified
10779 by the given coordinates.
10781 Default value is @samp{source}.
10785 Set when the expressions for coordinates @option{x0,y0,...x3,y3} are evaluated.
10787 It accepts the following values:
10790 only evaluate expressions once during the filter initialization or
10791 when a command is processed
10794 evaluate expressions for each incoming frame
10797 Default value is @samp{init}.
10802 Delay interlaced video by one field time so that the field order changes.
10804 The intended use is to fix PAL movies that have been captured with the
10805 opposite field order to the film-to-video transfer.
10807 A description of the accepted parameters follows.
10813 It accepts the following values:
10816 Capture field order top-first, transfer bottom-first.
10817 Filter will delay the bottom field.
10820 Capture field order bottom-first, transfer top-first.
10821 Filter will delay the top field.
10824 Capture and transfer with the same field order. This mode only exists
10825 for the documentation of the other options to refer to, but if you
10826 actually select it, the filter will faithfully do nothing.
10829 Capture field order determined automatically by field flags, transfer
10831 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
10832 basis using field flags. If no field information is available,
10833 then this works just like @samp{u}.
10836 Capture unknown or varying, transfer opposite.
10837 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
10838 analyzing the images and selecting the alternative that produces best
10839 match between the fields.
10842 Capture top-first, transfer unknown or varying.
10843 Filter selects among @samp{t} and @samp{p} using image analysis.
10846 Capture bottom-first, transfer unknown or varying.
10847 Filter selects among @samp{b} and @samp{p} using image analysis.
10850 Capture determined by field flags, transfer unknown or varying.
10851 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
10852 image analysis. If no field information is available, then this works just
10853 like @samp{U}. This is the default mode.
10856 Both capture and transfer unknown or varying.
10857 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
10861 @section pixdesctest
10863 Pixel format descriptor test filter, mainly useful for internal
10864 testing. The output video should be equal to the input video.
10868 format=monow, pixdesctest
10871 can be used to test the monowhite pixel format descriptor definition.
10875 Enable the specified chain of postprocessing subfilters using libpostproc. This
10876 library should be automatically selected with a GPL build (@code{--enable-gpl}).
10877 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
10878 Each subfilter and some options have a short and a long name that can be used
10879 interchangeably, i.e. dr/dering are the same.
10881 The filters accept the following options:
10885 Set postprocessing subfilters string.
10888 All subfilters share common options to determine their scope:
10892 Honor the quality commands for this subfilter.
10895 Do chrominance filtering, too (default).
10898 Do luminance filtering only (no chrominance).
10901 Do chrominance filtering only (no luminance).
10904 These options can be appended after the subfilter name, separated by a '|'.
10906 Available subfilters are:
10909 @item hb/hdeblock[|difference[|flatness]]
10910 Horizontal deblocking filter
10913 Difference factor where higher values mean more deblocking (default: @code{32}).
10915 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10918 @item vb/vdeblock[|difference[|flatness]]
10919 Vertical deblocking filter
10922 Difference factor where higher values mean more deblocking (default: @code{32}).
10924 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10927 @item ha/hadeblock[|difference[|flatness]]
10928 Accurate horizontal deblocking filter
10931 Difference factor where higher values mean more deblocking (default: @code{32}).
10933 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10936 @item va/vadeblock[|difference[|flatness]]
10937 Accurate vertical deblocking filter
10940 Difference factor where higher values mean more deblocking (default: @code{32}).
10942 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10946 The horizontal and vertical deblocking filters share the difference and
10947 flatness values so you cannot set different horizontal and vertical
10951 @item h1/x1hdeblock
10952 Experimental horizontal deblocking filter
10954 @item v1/x1vdeblock
10955 Experimental vertical deblocking filter
10960 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
10963 larger -> stronger filtering
10965 larger -> stronger filtering
10967 larger -> stronger filtering
10970 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
10973 Stretch luminance to @code{0-255}.
10976 @item lb/linblenddeint
10977 Linear blend deinterlacing filter that deinterlaces the given block by
10978 filtering all lines with a @code{(1 2 1)} filter.
10980 @item li/linipoldeint
10981 Linear interpolating deinterlacing filter that deinterlaces the given block by
10982 linearly interpolating every second line.
10984 @item ci/cubicipoldeint
10985 Cubic interpolating deinterlacing filter deinterlaces the given block by
10986 cubically interpolating every second line.
10988 @item md/mediandeint
10989 Median deinterlacing filter that deinterlaces the given block by applying a
10990 median filter to every second line.
10992 @item fd/ffmpegdeint
10993 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
10994 second line with a @code{(-1 4 2 4 -1)} filter.
10997 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
10998 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
11000 @item fq/forceQuant[|quantizer]
11001 Overrides the quantizer table from the input with the constant quantizer you
11009 Default pp filter combination (@code{hb|a,vb|a,dr|a})
11012 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
11015 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
11018 @subsection Examples
11022 Apply horizontal and vertical deblocking, deringing and automatic
11023 brightness/contrast:
11029 Apply default filters without brightness/contrast correction:
11035 Apply default filters and temporal denoiser:
11037 pp=default/tmpnoise|1|2|3
11041 Apply deblocking on luminance only, and switch vertical deblocking on or off
11042 automatically depending on available CPU time:
11049 Apply Postprocessing filter 7. It is variant of the @ref{spp} filter,
11050 similar to spp = 6 with 7 point DCT, where only the center sample is
11053 The filter accepts the following options:
11057 Force a constant quantization parameter. It accepts an integer in range
11058 0 to 63. If not set, the filter will use the QP from the video stream
11062 Set thresholding mode. Available modes are:
11066 Set hard thresholding.
11068 Set soft thresholding (better de-ringing effect, but likely blurrier).
11070 Set medium thresholding (good results, default).
11074 @section premultiply
11075 Apply alpha premultiply effect to input video stream using first plane
11076 of second stream as alpha.
11078 Both streams must have same dimensions and same pixel format.
11081 Apply prewitt operator to input video stream.
11083 The filter accepts the following option:
11087 Set which planes will be processed, unprocessed planes will be copied.
11088 By default value 0xf, all planes will be processed.
11091 Set value which will be multiplied with filtered result.
11094 Set value which will be added to filtered result.
11099 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
11100 Ratio) between two input videos.
11102 This filter takes in input two input videos, the first input is
11103 considered the "main" source and is passed unchanged to the
11104 output. The second input is used as a "reference" video for computing
11107 Both video inputs must have the same resolution and pixel format for
11108 this filter to work correctly. Also it assumes that both inputs
11109 have the same number of frames, which are compared one by one.
11111 The obtained average PSNR is printed through the logging system.
11113 The filter stores the accumulated MSE (mean squared error) of each
11114 frame, and at the end of the processing it is averaged across all frames
11115 equally, and the following formula is applied to obtain the PSNR:
11118 PSNR = 10*log10(MAX^2/MSE)
11121 Where MAX is the average of the maximum values of each component of the
11124 The description of the accepted parameters follows.
11127 @item stats_file, f
11128 If specified the filter will use the named file to save the PSNR of
11129 each individual frame. When filename equals "-" the data is sent to
11132 @item stats_version
11133 Specifies which version of the stats file format to use. Details of
11134 each format are written below.
11135 Default value is 1.
11137 @item stats_add_max
11138 Determines whether the max value is output to the stats log.
11139 Default value is 0.
11140 Requires stats_version >= 2. If this is set and stats_version < 2,
11141 the filter will return an error.
11144 The file printed if @var{stats_file} is selected, contains a sequence of
11145 key/value pairs of the form @var{key}:@var{value} for each compared
11148 If a @var{stats_version} greater than 1 is specified, a header line precedes
11149 the list of per-frame-pair stats, with key value pairs following the frame
11150 format with the following parameters:
11153 @item psnr_log_version
11154 The version of the log file format. Will match @var{stats_version}.
11157 A comma separated list of the per-frame-pair parameters included in
11161 A description of each shown per-frame-pair parameter follows:
11165 sequential number of the input frame, starting from 1
11168 Mean Square Error pixel-by-pixel average difference of the compared
11169 frames, averaged over all the image components.
11171 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
11172 Mean Square Error pixel-by-pixel average difference of the compared
11173 frames for the component specified by the suffix.
11175 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
11176 Peak Signal to Noise ratio of the compared frames for the component
11177 specified by the suffix.
11179 @item max_avg, max_y, max_u, max_v
11180 Maximum allowed value for each channel, and average over all
11186 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
11187 [main][ref] psnr="stats_file=stats.log" [out]
11190 On this example the input file being processed is compared with the
11191 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
11192 is stored in @file{stats.log}.
11197 Pulldown reversal (inverse telecine) filter, capable of handling mixed
11198 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
11201 The pullup filter is designed to take advantage of future context in making
11202 its decisions. This filter is stateless in the sense that it does not lock
11203 onto a pattern to follow, but it instead looks forward to the following
11204 fields in order to identify matches and rebuild progressive frames.
11206 To produce content with an even framerate, insert the fps filter after
11207 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
11208 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
11210 The filter accepts the following options:
11217 These options set the amount of "junk" to ignore at the left, right, top, and
11218 bottom of the image, respectively. Left and right are in units of 8 pixels,
11219 while top and bottom are in units of 2 lines.
11220 The default is 8 pixels on each side.
11223 Set the strict breaks. Setting this option to 1 will reduce the chances of
11224 filter generating an occasional mismatched frame, but it may also cause an
11225 excessive number of frames to be dropped during high motion sequences.
11226 Conversely, setting it to -1 will make filter match fields more easily.
11227 This may help processing of video where there is slight blurring between
11228 the fields, but may also cause there to be interlaced frames in the output.
11229 Default value is @code{0}.
11232 Set the metric plane to use. It accepts the following values:
11238 Use chroma blue plane.
11241 Use chroma red plane.
11244 This option may be set to use chroma plane instead of the default luma plane
11245 for doing filter's computations. This may improve accuracy on very clean
11246 source material, but more likely will decrease accuracy, especially if there
11247 is chroma noise (rainbow effect) or any grayscale video.
11248 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
11249 load and make pullup usable in realtime on slow machines.
11252 For best results (without duplicated frames in the output file) it is
11253 necessary to change the output frame rate. For example, to inverse
11254 telecine NTSC input:
11256 ffmpeg -i input -vf pullup -r 24000/1001 ...
11261 Change video quantization parameters (QP).
11263 The filter accepts the following option:
11267 Set expression for quantization parameter.
11270 The expression is evaluated through the eval API and can contain, among others,
11271 the following constants:
11275 1 if index is not 129, 0 otherwise.
11278 Sequentional index starting from -129 to 128.
11281 @subsection Examples
11285 Some equation like:
11293 Flush video frames from internal cache of frames into a random order.
11294 No frame is discarded.
11295 Inspired by @ref{frei0r} nervous filter.
11299 Set size in number of frames of internal cache, in range from @code{2} to
11300 @code{512}. Default is @code{30}.
11303 Set seed for random number generator, must be an integer included between
11304 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
11305 less than @code{0}, the filter will try to use a good random seed on a
11309 @section readeia608
11311 Read closed captioning (EIA-608) information from the top lines of a video frame.
11313 This filter adds frame metadata for @code{lavfi.readeia608.X.cc} and
11314 @code{lavfi.readeia608.X.line}, where @code{X} is the number of the identified line
11315 with EIA-608 data (starting from 0). A description of each metadata value follows:
11318 @item lavfi.readeia608.X.cc
11319 The two bytes stored as EIA-608 data (printed in hexadecimal).
11321 @item lavfi.readeia608.X.line
11322 The number of the line on which the EIA-608 data was identified and read.
11325 This filter accepts the following options:
11329 Set the line to start scanning for EIA-608 data. Default is @code{0}.
11332 Set the line to end scanning for EIA-608 data. Default is @code{29}.
11335 Set minimal acceptable amplitude change for sync codes detection.
11336 Default is @code{0.2}. Allowed range is @code{[0.001 - 1]}.
11339 Set the ratio of width reserved for sync code detection.
11340 Default is @code{0.27}. Allowed range is @code{[0.01 - 0.7]}.
11343 Set the max peaks height difference for sync code detection.
11344 Default is @code{0.1}. Allowed range is @code{[0.0 - 0.5]}.
11347 Set max peaks period difference for sync code detection.
11348 Default is @code{0.1}. Allowed range is @code{[0.0 - 0.5]}.
11351 Set the first two max start code bits differences.
11352 Default is @code{0.02}. Allowed range is @code{[0.0 - 0.5]}.
11355 Set the minimum ratio of bits height compared to 3rd start code bit.
11356 Default is @code{0.75}. Allowed range is @code{[0.01 - 1]}.
11359 Set the white color threshold. Default is @code{0.35}. Allowed range is @code{[0.1 - 1]}.
11362 Set the black color threshold. Default is @code{0.15}. Allowed range is @code{[0.0 - 0.5]}.
11365 Enable checking the parity bit. In the event of a parity error, the filter will output
11366 @code{0x00} for that character. Default is false.
11369 @subsection Examples
11373 Output a csv with presentation time and the first two lines of identified EIA-608 captioning data.
11375 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
11381 Read vertical interval timecode (VITC) information from the top lines of a
11384 The filter adds frame metadata key @code{lavfi.readvitc.tc_str} with the
11385 timecode value, if a valid timecode has been detected. Further metadata key
11386 @code{lavfi.readvitc.found} is set to 0/1 depending on whether
11387 timecode data has been found or not.
11389 This filter accepts the following options:
11393 Set the maximum number of lines to scan for VITC data. If the value is set to
11394 @code{-1} the full video frame is scanned. Default is @code{45}.
11397 Set the luma threshold for black. Accepts float numbers in the range [0.0,1.0],
11398 default value is @code{0.2}. The value must be equal or less than @code{thr_w}.
11401 Set the luma threshold for white. Accepts float numbers in the range [0.0,1.0],
11402 default value is @code{0.6}. The value must be equal or greater than @code{thr_b}.
11405 @subsection Examples
11409 Detect and draw VITC data onto the video frame; if no valid VITC is detected,
11410 draw @code{--:--:--:--} as a placeholder:
11412 ffmpeg -i input.avi -filter:v 'readvitc,drawtext=fontfile=FreeMono.ttf:text=%@{metadata\\:lavfi.readvitc.tc_str\\:--\\\\\\:--\\\\\\:--\\\\\\:--@}:x=(w-tw)/2:y=400-ascent'
11418 Remap pixels using 2nd: Xmap and 3rd: Ymap input video stream.
11420 Destination pixel at position (X, Y) will be picked from source (x, y) position
11421 where x = Xmap(X, Y) and y = Ymap(X, Y). If mapping values are out of range, zero
11422 value for pixel will be used for destination pixel.
11424 Xmap and Ymap input video streams must be of same dimensions. Output video stream
11425 will have Xmap/Ymap video stream dimensions.
11426 Xmap and Ymap input video streams are 16bit depth, single channel.
11428 @section removegrain
11430 The removegrain filter is a spatial denoiser for progressive video.
11434 Set mode for the first plane.
11437 Set mode for the second plane.
11440 Set mode for the third plane.
11443 Set mode for the fourth plane.
11446 Range of mode is from 0 to 24. Description of each mode follows:
11450 Leave input plane unchanged. Default.
11453 Clips the pixel with the minimum and maximum of the 8 neighbour pixels.
11456 Clips the pixel with the second minimum and maximum of the 8 neighbour pixels.
11459 Clips the pixel with the third minimum and maximum of the 8 neighbour pixels.
11462 Clips the pixel with the fourth minimum and maximum of the 8 neighbour pixels.
11463 This is equivalent to a median filter.
11466 Line-sensitive clipping giving the minimal change.
11469 Line-sensitive clipping, intermediate.
11472 Line-sensitive clipping, intermediate.
11475 Line-sensitive clipping, intermediate.
11478 Line-sensitive clipping on a line where the neighbours pixels are the closest.
11481 Replaces the target pixel with the closest neighbour.
11484 [1 2 1] horizontal and vertical kernel blur.
11490 Bob mode, interpolates top field from the line where the neighbours
11491 pixels are the closest.
11494 Bob mode, interpolates bottom field from the line where the neighbours
11495 pixels are the closest.
11498 Bob mode, interpolates top field. Same as 13 but with a more complicated
11499 interpolation formula.
11502 Bob mode, interpolates bottom field. Same as 14 but with a more complicated
11503 interpolation formula.
11506 Clips the pixel with the minimum and maximum of respectively the maximum and
11507 minimum of each pair of opposite neighbour pixels.
11510 Line-sensitive clipping using opposite neighbours whose greatest distance from
11511 the current pixel is minimal.
11514 Replaces the pixel with the average of its 8 neighbours.
11517 Averages the 9 pixels ([1 1 1] horizontal and vertical blur).
11520 Clips pixels using the averages of opposite neighbour.
11523 Same as mode 21 but simpler and faster.
11526 Small edge and halo removal, but reputed useless.
11532 @section removelogo
11534 Suppress a TV station logo, using an image file to determine which
11535 pixels comprise the logo. It works by filling in the pixels that
11536 comprise the logo with neighboring pixels.
11538 The filter accepts the following options:
11542 Set the filter bitmap file, which can be any image format supported by
11543 libavformat. The width and height of the image file must match those of the
11544 video stream being processed.
11547 Pixels in the provided bitmap image with a value of zero are not
11548 considered part of the logo, non-zero pixels are considered part of
11549 the logo. If you use white (255) for the logo and black (0) for the
11550 rest, you will be safe. For making the filter bitmap, it is
11551 recommended to take a screen capture of a black frame with the logo
11552 visible, and then using a threshold filter followed by the erode
11553 filter once or twice.
11555 If needed, little splotches can be fixed manually. Remember that if
11556 logo pixels are not covered, the filter quality will be much
11557 reduced. Marking too many pixels as part of the logo does not hurt as
11558 much, but it will increase the amount of blurring needed to cover over
11559 the image and will destroy more information than necessary, and extra
11560 pixels will slow things down on a large logo.
11562 @section repeatfields
11564 This filter uses the repeat_field flag from the Video ES headers and hard repeats
11565 fields based on its value.
11569 Reverse a video clip.
11571 Warning: This filter requires memory to buffer the entire clip, so trimming
11574 @subsection Examples
11578 Take the first 5 seconds of a clip, and reverse it.
11586 Rotate video by an arbitrary angle expressed in radians.
11588 The filter accepts the following options:
11590 A description of the optional parameters follows.
11593 Set an expression for the angle by which to rotate the input video
11594 clockwise, expressed as a number of radians. A negative value will
11595 result in a counter-clockwise rotation. By default it is set to "0".
11597 This expression is evaluated for each frame.
11600 Set the output width expression, default value is "iw".
11601 This expression is evaluated just once during configuration.
11604 Set the output height expression, default value is "ih".
11605 This expression is evaluated just once during configuration.
11608 Enable bilinear interpolation if set to 1, a value of 0 disables
11609 it. Default value is 1.
11612 Set the color used to fill the output area not covered by the rotated
11613 image. For the general syntax of this option, check the "Color" section in the
11614 ffmpeg-utils manual. If the special value "none" is selected then no
11615 background is printed (useful for example if the background is never shown).
11617 Default value is "black".
11620 The expressions for the angle and the output size can contain the
11621 following constants and functions:
11625 sequential number of the input frame, starting from 0. It is always NAN
11626 before the first frame is filtered.
11629 time in seconds of the input frame, it is set to 0 when the filter is
11630 configured. It is always NAN before the first frame is filtered.
11634 horizontal and vertical chroma subsample values. For example for the
11635 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11639 the input video width and height
11643 the output width and height, that is the size of the padded area as
11644 specified by the @var{width} and @var{height} expressions
11648 the minimal width/height required for completely containing the input
11649 video rotated by @var{a} radians.
11651 These are only available when computing the @option{out_w} and
11652 @option{out_h} expressions.
11655 @subsection Examples
11659 Rotate the input by PI/6 radians clockwise:
11665 Rotate the input by PI/6 radians counter-clockwise:
11671 Rotate the input by 45 degrees clockwise:
11677 Apply a constant rotation with period T, starting from an angle of PI/3:
11679 rotate=PI/3+2*PI*t/T
11683 Make the input video rotation oscillating with a period of T
11684 seconds and an amplitude of A radians:
11686 rotate=A*sin(2*PI/T*t)
11690 Rotate the video, output size is chosen so that the whole rotating
11691 input video is always completely contained in the output:
11693 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
11697 Rotate the video, reduce the output size so that no background is ever
11700 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
11704 @subsection Commands
11706 The filter supports the following commands:
11710 Set the angle expression.
11711 The command accepts the same syntax of the corresponding option.
11713 If the specified expression is not valid, it is kept at its current
11719 Apply Shape Adaptive Blur.
11721 The filter accepts the following options:
11724 @item luma_radius, lr
11725 Set luma blur filter strength, must be a value in range 0.1-4.0, default
11726 value is 1.0. A greater value will result in a more blurred image, and
11727 in slower processing.
11729 @item luma_pre_filter_radius, lpfr
11730 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
11733 @item luma_strength, ls
11734 Set luma maximum difference between pixels to still be considered, must
11735 be a value in the 0.1-100.0 range, default value is 1.0.
11737 @item chroma_radius, cr
11738 Set chroma blur filter strength, must be a value in range -0.9-4.0. A
11739 greater value will result in a more blurred image, and in slower
11742 @item chroma_pre_filter_radius, cpfr
11743 Set chroma pre-filter radius, must be a value in the -0.9-2.0 range.
11745 @item chroma_strength, cs
11746 Set chroma maximum difference between pixels to still be considered,
11747 must be a value in the -0.9-100.0 range.
11750 Each chroma option value, if not explicitly specified, is set to the
11751 corresponding luma option value.
11756 Scale (resize) the input video, using the libswscale library.
11758 The scale filter forces the output display aspect ratio to be the same
11759 of the input, by changing the output sample aspect ratio.
11761 If the input image format is different from the format requested by
11762 the next filter, the scale filter will convert the input to the
11765 @subsection Options
11766 The filter accepts the following options, or any of the options
11767 supported by the libswscale scaler.
11769 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
11770 the complete list of scaler options.
11775 Set the output video dimension expression. Default value is the input
11778 If the value is 0, the input width is used for the output.
11780 If one of the values is -1, the scale filter will use a value that
11781 maintains the aspect ratio of the input image, calculated from the
11782 other specified dimension. If both of them are -1, the input size is
11785 If one of the values is -n with n > 1, the scale filter will also use a value
11786 that maintains the aspect ratio of the input image, calculated from the other
11787 specified dimension. After that it will, however, make sure that the calculated
11788 dimension is divisible by n and adjust the value if necessary.
11790 See below for the list of accepted constants for use in the dimension
11794 Specify when to evaluate @var{width} and @var{height} expression. It accepts the following values:
11798 Only evaluate expressions once during the filter initialization or when a command is processed.
11801 Evaluate expressions for each incoming frame.
11805 Default value is @samp{init}.
11809 Set the interlacing mode. It accepts the following values:
11813 Force interlaced aware scaling.
11816 Do not apply interlaced scaling.
11819 Select interlaced aware scaling depending on whether the source frames
11820 are flagged as interlaced or not.
11823 Default value is @samp{0}.
11826 Set libswscale scaling flags. See
11827 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11828 complete list of values. If not explicitly specified the filter applies
11832 @item param0, param1
11833 Set libswscale input parameters for scaling algorithms that need them. See
11834 @ref{sws_params,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11835 complete documentation. If not explicitly specified the filter applies
11841 Set the video size. For the syntax of this option, check the
11842 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11844 @item in_color_matrix
11845 @item out_color_matrix
11846 Set in/output YCbCr color space type.
11848 This allows the autodetected value to be overridden as well as allows forcing
11849 a specific value used for the output and encoder.
11851 If not specified, the color space type depends on the pixel format.
11857 Choose automatically.
11860 Format conforming to International Telecommunication Union (ITU)
11861 Recommendation BT.709.
11864 Set color space conforming to the United States Federal Communications
11865 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
11868 Set color space conforming to:
11872 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
11875 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
11878 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
11883 Set color space conforming to SMPTE ST 240:1999.
11888 Set in/output YCbCr sample range.
11890 This allows the autodetected value to be overridden as well as allows forcing
11891 a specific value used for the output and encoder. If not specified, the
11892 range depends on the pixel format. Possible values:
11896 Choose automatically.
11899 Set full range (0-255 in case of 8-bit luma).
11902 Set "MPEG" range (16-235 in case of 8-bit luma).
11905 @item force_original_aspect_ratio
11906 Enable decreasing or increasing output video width or height if necessary to
11907 keep the original aspect ratio. Possible values:
11911 Scale the video as specified and disable this feature.
11914 The output video dimensions will automatically be decreased if needed.
11917 The output video dimensions will automatically be increased if needed.
11921 One useful instance of this option is that when you know a specific device's
11922 maximum allowed resolution, you can use this to limit the output video to
11923 that, while retaining the aspect ratio. For example, device A allows
11924 1280x720 playback, and your video is 1920x800. Using this option (set it to
11925 decrease) and specifying 1280x720 to the command line makes the output
11928 Please note that this is a different thing than specifying -1 for @option{w}
11929 or @option{h}, you still need to specify the output resolution for this option
11934 The values of the @option{w} and @option{h} options are expressions
11935 containing the following constants:
11940 The input width and height
11944 These are the same as @var{in_w} and @var{in_h}.
11948 The output (scaled) width and height
11952 These are the same as @var{out_w} and @var{out_h}
11955 The same as @var{iw} / @var{ih}
11958 input sample aspect ratio
11961 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
11965 horizontal and vertical input chroma subsample values. For example for the
11966 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11970 horizontal and vertical output chroma subsample values. For example for the
11971 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11974 @subsection Examples
11978 Scale the input video to a size of 200x100
11983 This is equivalent to:
11994 Specify a size abbreviation for the output size:
11999 which can also be written as:
12005 Scale the input to 2x:
12007 scale=w=2*iw:h=2*ih
12011 The above is the same as:
12013 scale=2*in_w:2*in_h
12017 Scale the input to 2x with forced interlaced scaling:
12019 scale=2*iw:2*ih:interl=1
12023 Scale the input to half size:
12025 scale=w=iw/2:h=ih/2
12029 Increase the width, and set the height to the same size:
12035 Seek Greek harmony:
12042 Increase the height, and set the width to 3/2 of the height:
12044 scale=w=3/2*oh:h=3/5*ih
12048 Increase the size, making the size a multiple of the chroma
12051 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
12055 Increase the width to a maximum of 500 pixels,
12056 keeping the same aspect ratio as the input:
12058 scale=w='min(500\, iw*3/2):h=-1'
12062 @subsection Commands
12064 This filter supports the following commands:
12068 Set the output video dimension expression.
12069 The command accepts the same syntax of the corresponding option.
12071 If the specified expression is not valid, it is kept at its current
12077 Use the NVIDIA Performance Primitives (libnpp) to perform scaling and/or pixel
12078 format conversion on CUDA video frames. Setting the output width and height
12079 works in the same way as for the @var{scale} filter.
12081 The following additional options are accepted:
12084 The pixel format of the output CUDA frames. If set to the string "same" (the
12085 default), the input format will be kept. Note that automatic format negotiation
12086 and conversion is not yet supported for hardware frames
12089 The interpolation algorithm used for resizing. One of the following:
12096 @item cubic2p_bspline
12097 2-parameter cubic (B=1, C=0)
12099 @item cubic2p_catmullrom
12100 2-parameter cubic (B=0, C=1/2)
12102 @item cubic2p_b05c03
12103 2-parameter cubic (B=1/2, C=3/10)
12115 Scale (resize) the input video, based on a reference video.
12117 See the scale filter for available options, scale2ref supports the same but
12118 uses the reference video instead of the main input as basis.
12120 @subsection Examples
12124 Scale a subtitle stream to match the main video in size before overlaying
12126 'scale2ref[b][a];[a][b]overlay'
12130 @anchor{selectivecolor}
12131 @section selectivecolor
12133 Adjust cyan, magenta, yellow and black (CMYK) to certain ranges of colors (such
12134 as "reds", "yellows", "greens", "cyans", ...). The adjustment range is defined
12135 by the "purity" of the color (that is, how saturated it already is).
12137 This filter is similar to the Adobe Photoshop Selective Color tool.
12139 The filter accepts the following options:
12142 @item correction_method
12143 Select color correction method.
12145 Available values are:
12148 Specified adjustments are applied "as-is" (added/subtracted to original pixel
12151 Specified adjustments are relative to the original component value.
12153 Default is @code{absolute}.
12155 Adjustments for red pixels (pixels where the red component is the maximum)
12157 Adjustments for yellow pixels (pixels where the blue component is the minimum)
12159 Adjustments for green pixels (pixels where the green component is the maximum)
12161 Adjustments for cyan pixels (pixels where the red component is the minimum)
12163 Adjustments for blue pixels (pixels where the blue component is the maximum)
12165 Adjustments for magenta pixels (pixels where the green component is the minimum)
12167 Adjustments for white pixels (pixels where all components are greater than 128)
12169 Adjustments for all pixels except pure black and pure white
12171 Adjustments for black pixels (pixels where all components are lesser than 128)
12173 Specify a Photoshop selective color file (@code{.asv}) to import the settings from.
12176 All the adjustment settings (@option{reds}, @option{yellows}, ...) accept up to
12177 4 space separated floating point adjustment values in the [-1,1] range,
12178 respectively to adjust the amount of cyan, magenta, yellow and black for the
12179 pixels of its range.
12181 @subsection Examples
12185 Increase cyan by 50% and reduce yellow by 33% in every green areas, and
12186 increase magenta by 27% in blue areas:
12188 selectivecolor=greens=.5 0 -.33 0:blues=0 .27
12192 Use a Photoshop selective color preset:
12194 selectivecolor=psfile=MySelectiveColorPresets/Misty.asv
12198 @anchor{separatefields}
12199 @section separatefields
12201 The @code{separatefields} takes a frame-based video input and splits
12202 each frame into its components fields, producing a new half height clip
12203 with twice the frame rate and twice the frame count.
12205 This filter use field-dominance information in frame to decide which
12206 of each pair of fields to place first in the output.
12207 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
12209 @section setdar, setsar
12211 The @code{setdar} filter sets the Display Aspect Ratio for the filter
12214 This is done by changing the specified Sample (aka Pixel) Aspect
12215 Ratio, according to the following equation:
12217 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
12220 Keep in mind that the @code{setdar} filter does not modify the pixel
12221 dimensions of the video frame. Also, the display aspect ratio set by
12222 this filter may be changed by later filters in the filterchain,
12223 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
12226 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
12227 the filter output video.
12229 Note that as a consequence of the application of this filter, the
12230 output display aspect ratio will change according to the equation
12233 Keep in mind that the sample aspect ratio set by the @code{setsar}
12234 filter may be changed by later filters in the filterchain, e.g. if
12235 another "setsar" or a "setdar" filter is applied.
12237 It accepts the following parameters:
12240 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
12241 Set the aspect ratio used by the filter.
12243 The parameter can be a floating point number string, an expression, or
12244 a string of the form @var{num}:@var{den}, where @var{num} and
12245 @var{den} are the numerator and denominator of the aspect ratio. If
12246 the parameter is not specified, it is assumed the value "0".
12247 In case the form "@var{num}:@var{den}" is used, the @code{:} character
12251 Set the maximum integer value to use for expressing numerator and
12252 denominator when reducing the expressed aspect ratio to a rational.
12253 Default value is @code{100}.
12257 The parameter @var{sar} is an expression containing
12258 the following constants:
12262 These are approximated values for the mathematical constants e
12263 (Euler's number), pi (Greek pi), and phi (the golden ratio).
12266 The input width and height.
12269 These are the same as @var{w} / @var{h}.
12272 The input sample aspect ratio.
12275 The input display aspect ratio. It is the same as
12276 (@var{w} / @var{h}) * @var{sar}.
12279 Horizontal and vertical chroma subsample values. For example, for the
12280 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
12283 @subsection Examples
12288 To change the display aspect ratio to 16:9, specify one of the following:
12295 To change the sample aspect ratio to 10:11, specify:
12301 To set a display aspect ratio of 16:9, and specify a maximum integer value of
12302 1000 in the aspect ratio reduction, use the command:
12304 setdar=ratio=16/9:max=1000
12312 Force field for the output video frame.
12314 The @code{setfield} filter marks the interlace type field for the
12315 output frames. It does not change the input frame, but only sets the
12316 corresponding property, which affects how the frame is treated by
12317 following filters (e.g. @code{fieldorder} or @code{yadif}).
12319 The filter accepts the following options:
12324 Available values are:
12328 Keep the same field property.
12331 Mark the frame as bottom-field-first.
12334 Mark the frame as top-field-first.
12337 Mark the frame as progressive.
12343 Show a line containing various information for each input video frame.
12344 The input video is not modified.
12346 The shown line contains a sequence of key/value pairs of the form
12347 @var{key}:@var{value}.
12349 The following values are shown in the output:
12353 The (sequential) number of the input frame, starting from 0.
12356 The Presentation TimeStamp of the input frame, expressed as a number of
12357 time base units. The time base unit depends on the filter input pad.
12360 The Presentation TimeStamp of the input frame, expressed as a number of
12364 The position of the frame in the input stream, or -1 if this information is
12365 unavailable and/or meaningless (for example in case of synthetic video).
12368 The pixel format name.
12371 The sample aspect ratio of the input frame, expressed in the form
12372 @var{num}/@var{den}.
12375 The size of the input frame. For the syntax of this option, check the
12376 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12379 The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
12380 for bottom field first).
12383 This is 1 if the frame is a key frame, 0 otherwise.
12386 The picture type of the input frame ("I" for an I-frame, "P" for a
12387 P-frame, "B" for a B-frame, or "?" for an unknown type).
12388 Also refer to the documentation of the @code{AVPictureType} enum and of
12389 the @code{av_get_picture_type_char} function defined in
12390 @file{libavutil/avutil.h}.
12393 The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
12395 @item plane_checksum
12396 The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
12397 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
12400 @section showpalette
12402 Displays the 256 colors palette of each frame. This filter is only relevant for
12403 @var{pal8} pixel format frames.
12405 It accepts the following option:
12409 Set the size of the box used to represent one palette color entry. Default is
12410 @code{30} (for a @code{30x30} pixel box).
12413 @section shuffleframes
12415 Reorder and/or duplicate and/or drop video frames.
12417 It accepts the following parameters:
12421 Set the destination indexes of input frames.
12422 This is space or '|' separated list of indexes that maps input frames to output
12423 frames. Number of indexes also sets maximal value that each index may have.
12424 '-1' index have special meaning and that is to drop frame.
12427 The first frame has the index 0. The default is to keep the input unchanged.
12429 @subsection Examples
12433 Swap second and third frame of every three frames of the input:
12435 ffmpeg -i INPUT -vf "shuffleframes=0 2 1" OUTPUT
12439 Swap 10th and 1st frame of every ten frames of the input:
12441 ffmpeg -i INPUT -vf "shuffleframes=9 1 2 3 4 5 6 7 8 0" OUTPUT
12445 @section shuffleplanes
12447 Reorder and/or duplicate video planes.
12449 It accepts the following parameters:
12454 The index of the input plane to be used as the first output plane.
12457 The index of the input plane to be used as the second output plane.
12460 The index of the input plane to be used as the third output plane.
12463 The index of the input plane to be used as the fourth output plane.
12467 The first plane has the index 0. The default is to keep the input unchanged.
12469 @subsection Examples
12473 Swap the second and third planes of the input:
12475 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
12479 @anchor{signalstats}
12480 @section signalstats
12481 Evaluate various visual metrics that assist in determining issues associated
12482 with the digitization of analog video media.
12484 By default the filter will log these metadata values:
12488 Display the minimal Y value contained within the input frame. Expressed in
12492 Display the Y value at the 10% percentile within the input frame. Expressed in
12496 Display the average Y value within the input frame. Expressed in range of
12500 Display the Y value at the 90% percentile within the input frame. Expressed in
12504 Display the maximum Y value contained within the input frame. Expressed in
12508 Display the minimal U value contained within the input frame. Expressed in
12512 Display the U value at the 10% percentile within the input frame. Expressed in
12516 Display the average U value within the input frame. Expressed in range of
12520 Display the U value at the 90% percentile within the input frame. Expressed in
12524 Display the maximum U value contained within the input frame. Expressed in
12528 Display the minimal V value contained within the input frame. Expressed in
12532 Display the V value at the 10% percentile within the input frame. Expressed in
12536 Display the average V value within the input frame. Expressed in range of
12540 Display the V value at the 90% percentile within the input frame. Expressed in
12544 Display the maximum V value contained within the input frame. Expressed in
12548 Display the minimal saturation value contained within the input frame.
12549 Expressed in range of [0-~181.02].
12552 Display the saturation value at the 10% percentile within the input frame.
12553 Expressed in range of [0-~181.02].
12556 Display the average saturation value within the input frame. Expressed in range
12560 Display the saturation value at the 90% percentile within the input frame.
12561 Expressed in range of [0-~181.02].
12564 Display the maximum saturation value contained within the input frame.
12565 Expressed in range of [0-~181.02].
12568 Display the median value for hue within the input frame. Expressed in range of
12572 Display the average value for hue within the input frame. Expressed in range of
12576 Display the average of sample value difference between all values of the Y
12577 plane in the current frame and corresponding values of the previous input frame.
12578 Expressed in range of [0-255].
12581 Display the average of sample value difference between all values of the U
12582 plane in the current frame and corresponding values of the previous input frame.
12583 Expressed in range of [0-255].
12586 Display the average of sample value difference between all values of the V
12587 plane in the current frame and corresponding values of the previous input frame.
12588 Expressed in range of [0-255].
12591 Display bit depth of Y plane in current frame.
12592 Expressed in range of [0-16].
12595 Display bit depth of U plane in current frame.
12596 Expressed in range of [0-16].
12599 Display bit depth of V plane in current frame.
12600 Expressed in range of [0-16].
12603 The filter accepts the following options:
12609 @option{stat} specify an additional form of image analysis.
12610 @option{out} output video with the specified type of pixel highlighted.
12612 Both options accept the following values:
12616 Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
12617 unlike the neighboring pixels of the same field. Examples of temporal outliers
12618 include the results of video dropouts, head clogs, or tape tracking issues.
12621 Identify @var{vertical line repetition}. Vertical line repetition includes
12622 similar rows of pixels within a frame. In born-digital video vertical line
12623 repetition is common, but this pattern is uncommon in video digitized from an
12624 analog source. When it occurs in video that results from the digitization of an
12625 analog source it can indicate concealment from a dropout compensator.
12628 Identify pixels that fall outside of legal broadcast range.
12632 Set the highlight color for the @option{out} option. The default color is
12636 @subsection Examples
12640 Output data of various video metrics:
12642 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
12646 Output specific data about the minimum and maximum values of the Y plane per frame:
12648 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
12652 Playback video while highlighting pixels that are outside of broadcast range in red.
12654 ffplay example.mov -vf signalstats="out=brng:color=red"
12658 Playback video with signalstats metadata drawn over the frame.
12660 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
12663 The contents of signalstat_drawtext.txt used in the command are:
12666 Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
12667 U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
12668 V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
12669 saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
12677 Calculates the MPEG-7 Video Signature. The filter can handle more than one
12678 input. In this case the matching between the inputs can be calculated additionally.
12679 The filter always passes through the first input. The signature of each stream can
12680 be written into a file.
12682 It accepts the following options:
12686 Enable or disable the matching process.
12688 Available values are:
12692 Disable the calculation of a matching (default).
12694 Calculate the matching for the whole video and output whether the whole video
12695 matches or only parts.
12697 Calculate only until a matching is found or the video ends. Should be faster in
12702 Set the number of inputs. The option value must be a non negative integer.
12703 Default value is 1.
12706 Set the path to which the output is written. If there is more than one input,
12707 the path must be a prototype, i.e. must contain %d or %0nd (where n is a positive
12708 integer), that will be replaced with the input number. If no filename is
12709 specified, no output will be written. This is the default.
12712 Choose the output format.
12714 Available values are:
12718 Use the specified binary representation (default).
12720 Use the specified xml representation.
12724 Set threshold to detect one word as similar. The option value must be an integer
12725 greater than zero. The default value is 9000.
12728 Set threshold to detect all words as similar. The option value must be an integer
12729 greater than zero. The default value is 60000.
12732 Set threshold to detect frames as similar. The option value must be an integer
12733 greater than zero. The default value is 116.
12736 Set the minimum length of a sequence in frames to recognize it as matching
12737 sequence. The option value must be a non negative integer value.
12738 The default value is 0.
12741 Set the minimum relation, that matching frames to all frames must have.
12742 The option value must be a double value between 0 and 1. The default value is 0.5.
12745 @subsection Examples
12749 To calculate the signature of an input video and store it in signature.bin:
12751 ffmpeg -i input.mkv -vf signature=filename=signature.bin -map 0:v -f null -
12755 To detect whether two videos match and store the signatures in XML format in
12756 signature0.xml and signature1.xml:
12758 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 -
12766 Blur the input video without impacting the outlines.
12768 It accepts the following options:
12771 @item luma_radius, lr
12772 Set the luma radius. The option value must be a float number in
12773 the range [0.1,5.0] that specifies the variance of the gaussian filter
12774 used to blur the image (slower if larger). Default value is 1.0.
12776 @item luma_strength, ls
12777 Set the luma strength. The option value must be a float number
12778 in the range [-1.0,1.0] that configures the blurring. A value included
12779 in [0.0,1.0] will blur the image whereas a value included in
12780 [-1.0,0.0] will sharpen the image. Default value is 1.0.
12782 @item luma_threshold, lt
12783 Set the luma threshold used as a coefficient to determine
12784 whether a pixel should be blurred or not. The option value must be an
12785 integer in the range [-30,30]. A value of 0 will filter all the image,
12786 a value included in [0,30] will filter flat areas and a value included
12787 in [-30,0] will filter edges. Default value is 0.
12789 @item chroma_radius, cr
12790 Set the chroma radius. The option value must be a float number in
12791 the range [0.1,5.0] that specifies the variance of the gaussian filter
12792 used to blur the image (slower if larger). Default value is @option{luma_radius}.
12794 @item chroma_strength, cs
12795 Set the chroma strength. The option value must be a float number
12796 in the range [-1.0,1.0] that configures the blurring. A value included
12797 in [0.0,1.0] will blur the image whereas a value included in
12798 [-1.0,0.0] will sharpen the image. Default value is @option{luma_strength}.
12800 @item chroma_threshold, ct
12801 Set the chroma threshold used as a coefficient to determine
12802 whether a pixel should be blurred or not. The option value must be an
12803 integer in the range [-30,30]. A value of 0 will filter all the image,
12804 a value included in [0,30] will filter flat areas and a value included
12805 in [-30,0] will filter edges. Default value is @option{luma_threshold}.
12808 If a chroma option is not explicitly set, the corresponding luma value
12813 Obtain the SSIM (Structural SImilarity Metric) between two input videos.
12815 This filter takes in input two input videos, the first input is
12816 considered the "main" source and is passed unchanged to the
12817 output. The second input is used as a "reference" video for computing
12820 Both video inputs must have the same resolution and pixel format for
12821 this filter to work correctly. Also it assumes that both inputs
12822 have the same number of frames, which are compared one by one.
12824 The filter stores the calculated SSIM of each frame.
12826 The description of the accepted parameters follows.
12829 @item stats_file, f
12830 If specified the filter will use the named file to save the SSIM of
12831 each individual frame. When filename equals "-" the data is sent to
12835 The file printed if @var{stats_file} is selected, contains a sequence of
12836 key/value pairs of the form @var{key}:@var{value} for each compared
12839 A description of each shown parameter follows:
12843 sequential number of the input frame, starting from 1
12845 @item Y, U, V, R, G, B
12846 SSIM of the compared frames for the component specified by the suffix.
12849 SSIM of the compared frames for the whole frame.
12852 Same as above but in dB representation.
12857 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
12858 [main][ref] ssim="stats_file=stats.log" [out]
12861 On this example the input file being processed is compared with the
12862 reference file @file{ref_movie.mpg}. The SSIM of each individual frame
12863 is stored in @file{stats.log}.
12865 Another example with both psnr and ssim at same time:
12867 ffmpeg -i main.mpg -i ref.mpg -lavfi "ssim;[0:v][1:v]psnr" -f null -
12872 Convert between different stereoscopic image formats.
12874 The filters accept the following options:
12878 Set stereoscopic image format of input.
12880 Available values for input image formats are:
12883 side by side parallel (left eye left, right eye right)
12886 side by side crosseye (right eye left, left eye right)
12889 side by side parallel with half width resolution
12890 (left eye left, right eye right)
12893 side by side crosseye with half width resolution
12894 (right eye left, left eye right)
12897 above-below (left eye above, right eye below)
12900 above-below (right eye above, left eye below)
12903 above-below with half height resolution
12904 (left eye above, right eye below)
12907 above-below with half height resolution
12908 (right eye above, left eye below)
12911 alternating frames (left eye first, right eye second)
12914 alternating frames (right eye first, left eye second)
12917 interleaved rows (left eye has top row, right eye starts on next row)
12920 interleaved rows (right eye has top row, left eye starts on next row)
12923 interleaved columns, left eye first
12926 interleaved columns, right eye first
12928 Default value is @samp{sbsl}.
12932 Set stereoscopic image format of output.
12936 side by side parallel (left eye left, right eye right)
12939 side by side crosseye (right eye left, left eye right)
12942 side by side parallel with half width resolution
12943 (left eye left, right eye right)
12946 side by side crosseye with half width resolution
12947 (right eye left, left eye right)
12950 above-below (left eye above, right eye below)
12953 above-below (right eye above, left eye below)
12956 above-below with half height resolution
12957 (left eye above, right eye below)
12960 above-below with half height resolution
12961 (right eye above, left eye below)
12964 alternating frames (left eye first, right eye second)
12967 alternating frames (right eye first, left eye second)
12970 interleaved rows (left eye has top row, right eye starts on next row)
12973 interleaved rows (right eye has top row, left eye starts on next row)
12976 anaglyph red/blue gray
12977 (red filter on left eye, blue filter on right eye)
12980 anaglyph red/green gray
12981 (red filter on left eye, green filter on right eye)
12984 anaglyph red/cyan gray
12985 (red filter on left eye, cyan filter on right eye)
12988 anaglyph red/cyan half colored
12989 (red filter on left eye, cyan filter on right eye)
12992 anaglyph red/cyan color
12993 (red filter on left eye, cyan filter on right eye)
12996 anaglyph red/cyan color optimized with the least squares projection of dubois
12997 (red filter on left eye, cyan filter on right eye)
13000 anaglyph green/magenta gray
13001 (green filter on left eye, magenta filter on right eye)
13004 anaglyph green/magenta half colored
13005 (green filter on left eye, magenta filter on right eye)
13008 anaglyph green/magenta colored
13009 (green filter on left eye, magenta filter on right eye)
13012 anaglyph green/magenta color optimized with the least squares projection of dubois
13013 (green filter on left eye, magenta filter on right eye)
13016 anaglyph yellow/blue gray
13017 (yellow filter on left eye, blue filter on right eye)
13020 anaglyph yellow/blue half colored
13021 (yellow filter on left eye, blue filter on right eye)
13024 anaglyph yellow/blue colored
13025 (yellow filter on left eye, blue filter on right eye)
13028 anaglyph yellow/blue color optimized with the least squares projection of dubois
13029 (yellow filter on left eye, blue filter on right eye)
13032 mono output (left eye only)
13035 mono output (right eye only)
13038 checkerboard, left eye first
13041 checkerboard, right eye first
13044 interleaved columns, left eye first
13047 interleaved columns, right eye first
13053 Default value is @samp{arcd}.
13056 @subsection Examples
13060 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
13066 Convert input video from above below (left eye above, right eye below) to side by side crosseye.
13072 @section streamselect, astreamselect
13073 Select video or audio streams.
13075 The filter accepts the following options:
13079 Set number of inputs. Default is 2.
13082 Set input indexes to remap to outputs.
13085 @subsection Commands
13087 The @code{streamselect} and @code{astreamselect} filter supports the following
13092 Set input indexes to remap to outputs.
13095 @subsection Examples
13099 Select first 5 seconds 1st stream and rest of time 2nd stream:
13101 sendcmd='5.0 streamselect map 1',streamselect=inputs=2:map=0
13105 Same as above, but for audio:
13107 asendcmd='5.0 astreamselect map 1',astreamselect=inputs=2:map=0
13112 Apply sobel operator to input video stream.
13114 The filter accepts the following option:
13118 Set which planes will be processed, unprocessed planes will be copied.
13119 By default value 0xf, all planes will be processed.
13122 Set value which will be multiplied with filtered result.
13125 Set value which will be added to filtered result.
13131 Apply a simple postprocessing filter that compresses and decompresses the image
13132 at several (or - in the case of @option{quality} level @code{6} - all) shifts
13133 and average the results.
13135 The filter accepts the following options:
13139 Set quality. This option defines the number of levels for averaging. It accepts
13140 an integer in the range 0-6. If set to @code{0}, the filter will have no
13141 effect. A value of @code{6} means the higher quality. For each increment of
13142 that value the speed drops by a factor of approximately 2. Default value is
13146 Force a constant quantization parameter. If not set, the filter will use the QP
13147 from the video stream (if available).
13150 Set thresholding mode. Available modes are:
13154 Set hard thresholding (default).
13156 Set soft thresholding (better de-ringing effect, but likely blurrier).
13159 @item use_bframe_qp
13160 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
13161 option may cause flicker since the B-Frames have often larger QP. Default is
13162 @code{0} (not enabled).
13168 Draw subtitles on top of input video using the libass library.
13170 To enable compilation of this filter you need to configure FFmpeg with
13171 @code{--enable-libass}. This filter also requires a build with libavcodec and
13172 libavformat to convert the passed subtitles file to ASS (Advanced Substation
13173 Alpha) subtitles format.
13175 The filter accepts the following options:
13179 Set the filename of the subtitle file to read. It must be specified.
13181 @item original_size
13182 Specify the size of the original video, the video for which the ASS file
13183 was composed. For the syntax of this option, check the
13184 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13185 Due to a misdesign in ASS aspect ratio arithmetic, this is necessary to
13186 correctly scale the fonts if the aspect ratio has been changed.
13189 Set a directory path containing fonts that can be used by the filter.
13190 These fonts will be used in addition to whatever the font provider uses.
13193 Set subtitles input character encoding. @code{subtitles} filter only. Only
13194 useful if not UTF-8.
13196 @item stream_index, si
13197 Set subtitles stream index. @code{subtitles} filter only.
13200 Override default style or script info parameters of the subtitles. It accepts a
13201 string containing ASS style format @code{KEY=VALUE} couples separated by ",".
13204 If the first key is not specified, it is assumed that the first value
13205 specifies the @option{filename}.
13207 For example, to render the file @file{sub.srt} on top of the input
13208 video, use the command:
13213 which is equivalent to:
13215 subtitles=filename=sub.srt
13218 To render the default subtitles stream from file @file{video.mkv}, use:
13220 subtitles=video.mkv
13223 To render the second subtitles stream from that file, use:
13225 subtitles=video.mkv:si=1
13228 To make the subtitles stream from @file{sub.srt} appear in transparent green
13229 @code{DejaVu Serif}, use:
13231 subtitles=sub.srt:force_style='FontName=DejaVu Serif,PrimaryColour=&HAA00FF00'
13234 @section super2xsai
13236 Scale the input by 2x and smooth using the Super2xSaI (Scale and
13237 Interpolate) pixel art scaling algorithm.
13239 Useful for enlarging pixel art images without reducing sharpness.
13243 Swap two rectangular objects in video.
13245 This filter accepts the following options:
13255 Set 1st rect x coordinate.
13258 Set 1st rect y coordinate.
13261 Set 2nd rect x coordinate.
13264 Set 2nd rect y coordinate.
13266 All expressions are evaluated once for each frame.
13269 The all options are expressions containing the following constants:
13274 The input width and height.
13277 same as @var{w} / @var{h}
13280 input sample aspect ratio
13283 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
13286 The number of the input frame, starting from 0.
13289 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
13292 the position in the file of the input frame, NAN if unknown
13300 Apply telecine process to the video.
13302 This filter accepts the following options:
13311 The default value is @code{top}.
13315 A string of numbers representing the pulldown pattern you wish to apply.
13316 The default value is @code{23}.
13320 Some typical patterns:
13325 24p: 2332 (preferred)
13332 24p: 222222222223 ("Euro pulldown")
13339 Apply threshold effect to video stream.
13341 This filter needs four video streams to perform thresholding.
13342 First stream is stream we are filtering.
13343 Second stream is holding threshold values, third stream is holding min values,
13344 and last, fourth stream is holding max values.
13346 The filter accepts the following option:
13350 Set which planes will be processed, unprocessed planes will be copied.
13351 By default value 0xf, all planes will be processed.
13354 For example if first stream pixel's component value is less then threshold value
13355 of pixel component from 2nd threshold stream, third stream value will picked,
13356 otherwise fourth stream pixel component value will be picked.
13358 Using color source filter one can perform various types of thresholding:
13360 @subsection Examples
13364 Binary threshold, using gray color as threshold:
13366 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=black -f lavfi -i color=white -lavfi threshold output.avi
13370 Inverted binary threshold, using gray color as threshold:
13372 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -f lavfi -i color=black -lavfi threshold output.avi
13376 Truncate binary threshold, using gray color as threshold:
13378 ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=gray -lavfi threshold output.avi
13382 Threshold to zero, using gray color as threshold:
13384 ffmpeg -i 320x240.avi -f lavfi -i color=gray -f lavfi -i color=white -i 320x240.avi -lavfi threshold output.avi
13388 Inverted threshold to zero, using gray color as threshold:
13390 ffmpeg -i 320x240.avi -f lavfi -i color=gray -i 320x240.avi -f lavfi -i color=white -lavfi threshold output.avi
13395 Select the most representative frame in a given sequence of consecutive frames.
13397 The filter accepts the following options:
13401 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
13402 will pick one of them, and then handle the next batch of @var{n} frames until
13403 the end. Default is @code{100}.
13406 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
13407 value will result in a higher memory usage, so a high value is not recommended.
13409 @subsection Examples
13413 Extract one picture each 50 frames:
13419 Complete example of a thumbnail creation with @command{ffmpeg}:
13421 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
13427 Tile several successive frames together.
13429 The filter accepts the following options:
13434 Set the grid size (i.e. the number of lines and columns). For the syntax of
13435 this option, check the
13436 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13439 Set the maximum number of frames to render in the given area. It must be less
13440 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
13441 the area will be used.
13444 Set the outer border margin in pixels.
13447 Set the inner border thickness (i.e. the number of pixels between frames). For
13448 more advanced padding options (such as having different values for the edges),
13449 refer to the pad video filter.
13452 Specify the color of the unused area. For the syntax of this option, check the
13453 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
13457 @subsection Examples
13461 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
13463 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
13465 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
13466 duplicating each output frame to accommodate the originally detected frame
13470 Display @code{5} pictures in an area of @code{3x2} frames,
13471 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
13472 mixed flat and named options:
13474 tile=3x2:nb_frames=5:padding=7:margin=2
13478 @section tinterlace
13480 Perform various types of temporal field interlacing.
13482 Frames are counted starting from 1, so the first input frame is
13485 The filter accepts the following options:
13490 Specify the mode of the interlacing. This option can also be specified
13491 as a value alone. See below for a list of values for this option.
13493 Available values are:
13497 Move odd frames into the upper field, even into the lower field,
13498 generating a double height frame at half frame rate.
13502 Frame 1 Frame 2 Frame 3 Frame 4
13504 11111 22222 33333 44444
13505 11111 22222 33333 44444
13506 11111 22222 33333 44444
13507 11111 22222 33333 44444
13521 Only output odd frames, even frames are dropped, generating a frame with
13522 unchanged height at half frame rate.
13527 Frame 1 Frame 2 Frame 3 Frame 4
13529 11111 22222 33333 44444
13530 11111 22222 33333 44444
13531 11111 22222 33333 44444
13532 11111 22222 33333 44444
13542 Only output even frames, odd frames are dropped, generating a frame with
13543 unchanged height at half frame rate.
13548 Frame 1 Frame 2 Frame 3 Frame 4
13550 11111 22222 33333 44444
13551 11111 22222 33333 44444
13552 11111 22222 33333 44444
13553 11111 22222 33333 44444
13563 Expand each frame to full height, but pad alternate lines with black,
13564 generating a frame with double height at the same input frame rate.
13569 Frame 1 Frame 2 Frame 3 Frame 4
13571 11111 22222 33333 44444
13572 11111 22222 33333 44444
13573 11111 22222 33333 44444
13574 11111 22222 33333 44444
13577 11111 ..... 33333 .....
13578 ..... 22222 ..... 44444
13579 11111 ..... 33333 .....
13580 ..... 22222 ..... 44444
13581 11111 ..... 33333 .....
13582 ..... 22222 ..... 44444
13583 11111 ..... 33333 .....
13584 ..... 22222 ..... 44444
13588 @item interleave_top, 4
13589 Interleave the upper field from odd frames with the lower field from
13590 even frames, generating a frame with unchanged height at half frame rate.
13595 Frame 1 Frame 2 Frame 3 Frame 4
13597 11111<- 22222 33333<- 44444
13598 11111 22222<- 33333 44444<-
13599 11111<- 22222 33333<- 44444
13600 11111 22222<- 33333 44444<-
13610 @item interleave_bottom, 5
13611 Interleave the lower field from odd frames with the upper field from
13612 even frames, generating a frame with unchanged height at half frame rate.
13617 Frame 1 Frame 2 Frame 3 Frame 4
13619 11111 22222<- 33333 44444<-
13620 11111<- 22222 33333<- 44444
13621 11111 22222<- 33333 44444<-
13622 11111<- 22222 33333<- 44444
13632 @item interlacex2, 6
13633 Double frame rate with unchanged height. Frames are inserted each
13634 containing the second temporal field from the previous input frame and
13635 the first temporal field from the next input frame. This mode relies on
13636 the top_field_first flag. Useful for interlaced video displays with no
13637 field synchronisation.
13642 Frame 1 Frame 2 Frame 3 Frame 4
13644 11111 22222 33333 44444
13645 11111 22222 33333 44444
13646 11111 22222 33333 44444
13647 11111 22222 33333 44444
13650 11111 22222 22222 33333 33333 44444 44444
13651 11111 11111 22222 22222 33333 33333 44444
13652 11111 22222 22222 33333 33333 44444 44444
13653 11111 11111 22222 22222 33333 33333 44444
13658 Move odd frames into the upper field, even into the lower field,
13659 generating a double height frame at same frame rate.
13664 Frame 1 Frame 2 Frame 3 Frame 4
13666 11111 22222 33333 44444
13667 11111 22222 33333 44444
13668 11111 22222 33333 44444
13669 11111 22222 33333 44444
13672 11111 33333 33333 55555
13673 22222 22222 44444 44444
13674 11111 33333 33333 55555
13675 22222 22222 44444 44444
13676 11111 33333 33333 55555
13677 22222 22222 44444 44444
13678 11111 33333 33333 55555
13679 22222 22222 44444 44444
13684 Numeric values are deprecated but are accepted for backward
13685 compatibility reasons.
13687 Default mode is @code{merge}.
13690 Specify flags influencing the filter process.
13692 Available value for @var{flags} is:
13695 @item low_pass_filter, vlfp
13696 Enable vertical low-pass filtering in the filter.
13697 Vertical low-pass filtering is required when creating an interlaced
13698 destination from a progressive source which contains high-frequency
13699 vertical detail. Filtering will reduce interlace 'twitter' and Moire
13702 Vertical low-pass filtering can only be enabled for @option{mode}
13703 @var{interleave_top} and @var{interleave_bottom}.
13710 Transpose rows with columns in the input video and optionally flip it.
13712 It accepts the following parameters:
13717 Specify the transposition direction.
13719 Can assume the following values:
13721 @item 0, 4, cclock_flip
13722 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
13730 Rotate by 90 degrees clockwise, that is:
13738 Rotate by 90 degrees counterclockwise, that is:
13745 @item 3, 7, clock_flip
13746 Rotate by 90 degrees clockwise and vertically flip, that is:
13754 For values between 4-7, the transposition is only done if the input
13755 video geometry is portrait and not landscape. These values are
13756 deprecated, the @code{passthrough} option should be used instead.
13758 Numerical values are deprecated, and should be dropped in favor of
13759 symbolic constants.
13762 Do not apply the transposition if the input geometry matches the one
13763 specified by the specified value. It accepts the following values:
13766 Always apply transposition.
13768 Preserve portrait geometry (when @var{height} >= @var{width}).
13770 Preserve landscape geometry (when @var{width} >= @var{height}).
13773 Default value is @code{none}.
13776 For example to rotate by 90 degrees clockwise and preserve portrait
13779 transpose=dir=1:passthrough=portrait
13782 The command above can also be specified as:
13784 transpose=1:portrait
13788 Trim the input so that the output contains one continuous subpart of the input.
13790 It accepts the following parameters:
13793 Specify the time of the start of the kept section, i.e. the frame with the
13794 timestamp @var{start} will be the first frame in the output.
13797 Specify the time of the first frame that will be dropped, i.e. the frame
13798 immediately preceding the one with the timestamp @var{end} will be the last
13799 frame in the output.
13802 This is the same as @var{start}, except this option sets the start timestamp
13803 in timebase units instead of seconds.
13806 This is the same as @var{end}, except this option sets the end timestamp
13807 in timebase units instead of seconds.
13810 The maximum duration of the output in seconds.
13813 The number of the first frame that should be passed to the output.
13816 The number of the first frame that should be dropped.
13819 @option{start}, @option{end}, and @option{duration} are expressed as time
13820 duration specifications; see
13821 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
13822 for the accepted syntax.
13824 Note that the first two sets of the start/end options and the @option{duration}
13825 option look at the frame timestamp, while the _frame variants simply count the
13826 frames that pass through the filter. Also note that this filter does not modify
13827 the timestamps. If you wish for the output timestamps to start at zero, insert a
13828 setpts filter after the trim filter.
13830 If multiple start or end options are set, this filter tries to be greedy and
13831 keep all the frames that match at least one of the specified constraints. To keep
13832 only the part that matches all the constraints at once, chain multiple trim
13835 The defaults are such that all the input is kept. So it is possible to set e.g.
13836 just the end values to keep everything before the specified time.
13841 Drop everything except the second minute of input:
13843 ffmpeg -i INPUT -vf trim=60:120
13847 Keep only the first second:
13849 ffmpeg -i INPUT -vf trim=duration=1
13858 Sharpen or blur the input video.
13860 It accepts the following parameters:
13863 @item luma_msize_x, lx
13864 Set the luma matrix horizontal size. It must be an odd integer between
13865 3 and 23. The default value is 5.
13867 @item luma_msize_y, ly
13868 Set the luma matrix vertical size. It must be an odd integer between 3
13869 and 23. The default value is 5.
13871 @item luma_amount, la
13872 Set the luma effect strength. It must be a floating point number, reasonable
13873 values lay between -1.5 and 1.5.
13875 Negative values will blur the input video, while positive values will
13876 sharpen it, a value of zero will disable the effect.
13878 Default value is 1.0.
13880 @item chroma_msize_x, cx
13881 Set the chroma matrix horizontal size. It must be an odd integer
13882 between 3 and 23. The default value is 5.
13884 @item chroma_msize_y, cy
13885 Set the chroma matrix vertical size. It must be an odd integer
13886 between 3 and 23. The default value is 5.
13888 @item chroma_amount, ca
13889 Set the chroma effect strength. It must be a floating point number, reasonable
13890 values lay between -1.5 and 1.5.
13892 Negative values will blur the input video, while positive values will
13893 sharpen it, a value of zero will disable the effect.
13895 Default value is 0.0.
13898 If set to 1, specify using OpenCL capabilities, only available if
13899 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
13903 All parameters are optional and default to the equivalent of the
13904 string '5:5:1.0:5:5:0.0'.
13906 @subsection Examples
13910 Apply strong luma sharpen effect:
13912 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
13916 Apply a strong blur of both luma and chroma parameters:
13918 unsharp=7:7:-2:7:7:-2
13924 Apply ultra slow/simple postprocessing filter that compresses and decompresses
13925 the image at several (or - in the case of @option{quality} level @code{8} - all)
13926 shifts and average the results.
13928 The way this differs from the behavior of spp is that uspp actually encodes &
13929 decodes each case with libavcodec Snow, whereas spp uses a simplified intra only 8x8
13930 DCT similar to MJPEG.
13932 The filter accepts the following options:
13936 Set quality. This option defines the number of levels for averaging. It accepts
13937 an integer in the range 0-8. If set to @code{0}, the filter will have no
13938 effect. A value of @code{8} means the higher quality. For each increment of
13939 that value the speed drops by a factor of approximately 2. Default value is
13943 Force a constant quantization parameter. If not set, the filter will use the QP
13944 from the video stream (if available).
13947 @section vaguedenoiser
13949 Apply a wavelet based denoiser.
13951 It transforms each frame from the video input into the wavelet domain,
13952 using Cohen-Daubechies-Feauveau 9/7. Then it applies some filtering to
13953 the obtained coefficients. It does an inverse wavelet transform after.
13954 Due to wavelet properties, it should give a nice smoothed result, and
13955 reduced noise, without blurring picture features.
13957 This filter accepts the following options:
13961 The filtering strength. The higher, the more filtered the video will be.
13962 Hard thresholding can use a higher threshold than soft thresholding
13963 before the video looks overfiltered.
13966 The filtering method the filter will use.
13968 It accepts the following values:
13971 All values under the threshold will be zeroed.
13974 All values under the threshold will be zeroed. All values above will be
13975 reduced by the threshold.
13978 Scales or nullifies coefficients - intermediary between (more) soft and
13979 (less) hard thresholding.
13983 Number of times, the wavelet will decompose the picture. Picture can't
13984 be decomposed beyond a particular point (typically, 8 for a 640x480
13985 frame - as 2^9 = 512 > 480)
13988 Partial of full denoising (limited coefficients shrinking), from 0 to 100.
13991 A list of the planes to process. By default all planes are processed.
13994 @section vectorscope
13996 Display 2 color component values in the two dimensional graph (which is called
13999 This filter accepts the following options:
14003 Set vectorscope mode.
14005 It accepts the following values:
14008 Gray values are displayed on graph, higher brightness means more pixels have
14009 same component color value on location in graph. This is the default mode.
14012 Gray values are displayed on graph. Surrounding pixels values which are not
14013 present in video frame are drawn in gradient of 2 color components which are
14014 set by option @code{x} and @code{y}. The 3rd color component is static.
14017 Actual color components values present in video frame are displayed on graph.
14020 Similar as color2 but higher frequency of same values @code{x} and @code{y}
14021 on graph increases value of another color component, which is luminance by
14022 default values of @code{x} and @code{y}.
14025 Actual colors present in video frame are displayed on graph. If two different
14026 colors map to same position on graph then color with higher value of component
14027 not present in graph is picked.
14030 Gray values are displayed on graph. Similar to @code{color} but with 3rd color
14031 component picked from radial gradient.
14035 Set which color component will be represented on X-axis. Default is @code{1}.
14038 Set which color component will be represented on Y-axis. Default is @code{2}.
14041 Set intensity, used by modes: gray, color, color3 and color5 for increasing brightness
14042 of color component which represents frequency of (X, Y) location in graph.
14047 No envelope, this is default.
14050 Instant envelope, even darkest single pixel will be clearly highlighted.
14053 Hold maximum and minimum values presented in graph over time. This way you
14054 can still spot out of range values without constantly looking at vectorscope.
14057 Peak and instant envelope combined together.
14061 Set what kind of graticule to draw.
14069 Set graticule opacity.
14072 Set graticule flags.
14076 Draw graticule for white point.
14079 Draw graticule for black point.
14082 Draw color points short names.
14086 Set background opacity.
14088 @item lthreshold, l
14089 Set low threshold for color component not represented on X or Y axis.
14090 Values lower than this value will be ignored. Default is 0.
14091 Note this value is multiplied with actual max possible value one pixel component
14092 can have. So for 8-bit input and low threshold value of 0.1 actual threshold
14095 @item hthreshold, h
14096 Set high threshold for color component not represented on X or Y axis.
14097 Values higher than this value will be ignored. Default is 1.
14098 Note this value is multiplied with actual max possible value one pixel component
14099 can have. So for 8-bit input and high threshold value of 0.9 actual threshold
14100 is 0.9 * 255 = 230.
14102 @item colorspace, c
14103 Set what kind of colorspace to use when drawing graticule.
14112 @anchor{vidstabdetect}
14113 @section vidstabdetect
14115 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
14116 @ref{vidstabtransform} for pass 2.
14118 This filter generates a file with relative translation and rotation
14119 transform information about subsequent frames, which is then used by
14120 the @ref{vidstabtransform} filter.
14122 To enable compilation of this filter you need to configure FFmpeg with
14123 @code{--enable-libvidstab}.
14125 This filter accepts the following options:
14129 Set the path to the file used to write the transforms information.
14130 Default value is @file{transforms.trf}.
14133 Set how shaky the video is and how quick the camera is. It accepts an
14134 integer in the range 1-10, a value of 1 means little shakiness, a
14135 value of 10 means strong shakiness. Default value is 5.
14138 Set the accuracy of the detection process. It must be a value in the
14139 range 1-15. A value of 1 means low accuracy, a value of 15 means high
14140 accuracy. Default value is 15.
14143 Set stepsize of the search process. The region around minimum is
14144 scanned with 1 pixel resolution. Default value is 6.
14147 Set minimum contrast. Below this value a local measurement field is
14148 discarded. Must be a floating point value in the range 0-1. Default
14152 Set reference frame number for tripod mode.
14154 If enabled, the motion of the frames is compared to a reference frame
14155 in the filtered stream, identified by the specified number. The idea
14156 is to compensate all movements in a more-or-less static scene and keep
14157 the camera view absolutely still.
14159 If set to 0, it is disabled. The frames are counted starting from 1.
14162 Show fields and transforms in the resulting frames. It accepts an
14163 integer in the range 0-2. Default value is 0, which disables any
14167 @subsection Examples
14171 Use default values:
14177 Analyze strongly shaky movie and put the results in file
14178 @file{mytransforms.trf}:
14180 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
14184 Visualize the result of internal transformations in the resulting
14187 vidstabdetect=show=1
14191 Analyze a video with medium shakiness using @command{ffmpeg}:
14193 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
14197 @anchor{vidstabtransform}
14198 @section vidstabtransform
14200 Video stabilization/deshaking: pass 2 of 2,
14201 see @ref{vidstabdetect} for pass 1.
14203 Read a file with transform information for each frame and
14204 apply/compensate them. Together with the @ref{vidstabdetect}
14205 filter this can be used to deshake videos. See also
14206 @url{http://public.hronopik.de/vid.stab}. It is important to also use
14207 the @ref{unsharp} filter, see below.
14209 To enable compilation of this filter you need to configure FFmpeg with
14210 @code{--enable-libvidstab}.
14212 @subsection Options
14216 Set path to the file used to read the transforms. Default value is
14217 @file{transforms.trf}.
14220 Set the number of frames (value*2 + 1) used for lowpass filtering the
14221 camera movements. Default value is 10.
14223 For example a number of 10 means that 21 frames are used (10 in the
14224 past and 10 in the future) to smoothen the motion in the video. A
14225 larger value leads to a smoother video, but limits the acceleration of
14226 the camera (pan/tilt movements). 0 is a special case where a static
14227 camera is simulated.
14230 Set the camera path optimization algorithm.
14232 Accepted values are:
14235 gaussian kernel low-pass filter on camera motion (default)
14237 averaging on transformations
14241 Set maximal number of pixels to translate frames. Default value is -1,
14245 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
14246 value is -1, meaning no limit.
14249 Specify how to deal with borders that may be visible due to movement
14252 Available values are:
14255 keep image information from previous frame (default)
14257 fill the border black
14261 Invert transforms if set to 1. Default value is 0.
14264 Consider transforms as relative to previous frame if set to 1,
14265 absolute if set to 0. Default value is 0.
14268 Set percentage to zoom. A positive value will result in a zoom-in
14269 effect, a negative value in a zoom-out effect. Default value is 0 (no
14273 Set optimal zooming to avoid borders.
14275 Accepted values are:
14280 optimal static zoom value is determined (only very strong movements
14281 will lead to visible borders) (default)
14283 optimal adaptive zoom value is determined (no borders will be
14284 visible), see @option{zoomspeed}
14287 Note that the value given at zoom is added to the one calculated here.
14290 Set percent to zoom maximally each frame (enabled when
14291 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
14295 Specify type of interpolation.
14297 Available values are:
14302 linear only horizontal
14304 linear in both directions (default)
14306 cubic in both directions (slow)
14310 Enable virtual tripod mode if set to 1, which is equivalent to
14311 @code{relative=0:smoothing=0}. Default value is 0.
14313 Use also @code{tripod} option of @ref{vidstabdetect}.
14316 Increase log verbosity if set to 1. Also the detected global motions
14317 are written to the temporary file @file{global_motions.trf}. Default
14321 @subsection Examples
14325 Use @command{ffmpeg} for a typical stabilization with default values:
14327 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
14330 Note the use of the @ref{unsharp} filter which is always recommended.
14333 Zoom in a bit more and load transform data from a given file:
14335 vidstabtransform=zoom=5:input="mytransforms.trf"
14339 Smoothen the video even more:
14341 vidstabtransform=smoothing=30
14347 Flip the input video vertically.
14349 For example, to vertically flip a video with @command{ffmpeg}:
14351 ffmpeg -i in.avi -vf "vflip" out.avi
14357 Make or reverse a natural vignetting effect.
14359 The filter accepts the following options:
14363 Set lens angle expression as a number of radians.
14365 The value is clipped in the @code{[0,PI/2]} range.
14367 Default value: @code{"PI/5"}
14371 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
14375 Set forward/backward mode.
14377 Available modes are:
14380 The larger the distance from the central point, the darker the image becomes.
14383 The larger the distance from the central point, the brighter the image becomes.
14384 This can be used to reverse a vignette effect, though there is no automatic
14385 detection to extract the lens @option{angle} and other settings (yet). It can
14386 also be used to create a burning effect.
14389 Default value is @samp{forward}.
14392 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
14394 It accepts the following values:
14397 Evaluate expressions only once during the filter initialization.
14400 Evaluate expressions for each incoming frame. This is way slower than the
14401 @samp{init} mode since it requires all the scalers to be re-computed, but it
14402 allows advanced dynamic expressions.
14405 Default value is @samp{init}.
14408 Set dithering to reduce the circular banding effects. Default is @code{1}
14412 Set vignette aspect. This setting allows one to adjust the shape of the vignette.
14413 Setting this value to the SAR of the input will make a rectangular vignetting
14414 following the dimensions of the video.
14416 Default is @code{1/1}.
14419 @subsection Expressions
14421 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
14422 following parameters.
14427 input width and height
14430 the number of input frame, starting from 0
14433 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
14434 @var{TB} units, NAN if undefined
14437 frame rate of the input video, NAN if the input frame rate is unknown
14440 the PTS (Presentation TimeStamp) of the filtered video frame,
14441 expressed in seconds, NAN if undefined
14444 time base of the input video
14448 @subsection Examples
14452 Apply simple strong vignetting effect:
14458 Make a flickering vignetting:
14460 vignette='PI/4+random(1)*PI/50':eval=frame
14466 Stack input videos vertically.
14468 All streams must be of same pixel format and of same width.
14470 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
14471 to create same output.
14473 The filter accept the following option:
14477 Set number of input streams. Default is 2.
14480 If set to 1, force the output to terminate when the shortest input
14481 terminates. Default value is 0.
14486 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
14487 Deinterlacing Filter").
14489 Based on the process described by Martin Weston for BBC R&D, and
14490 implemented based on the de-interlace algorithm written by Jim
14491 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
14492 uses filter coefficients calculated by BBC R&D.
14494 There are two sets of filter coefficients, so called "simple":
14495 and "complex". Which set of filter coefficients is used can
14496 be set by passing an optional parameter:
14500 Set the interlacing filter coefficients. Accepts one of the following values:
14504 Simple filter coefficient set.
14506 More-complex filter coefficient set.
14508 Default value is @samp{complex}.
14511 Specify which frames to deinterlace. Accept one of the following values:
14515 Deinterlace all frames,
14517 Only deinterlace frames marked as interlaced.
14520 Default value is @samp{all}.
14524 Video waveform monitor.
14526 The waveform monitor plots color component intensity. By default luminance
14527 only. Each column of the waveform corresponds to a column of pixels in the
14530 It accepts the following options:
14534 Can be either @code{row}, or @code{column}. Default is @code{column}.
14535 In row mode, the graph on the left side represents color component value 0 and
14536 the right side represents value = 255. In column mode, the top side represents
14537 color component value = 0 and bottom side represents value = 255.
14540 Set intensity. Smaller values are useful to find out how many values of the same
14541 luminance are distributed across input rows/columns.
14542 Default value is @code{0.04}. Allowed range is [0, 1].
14545 Set mirroring mode. @code{0} means unmirrored, @code{1} means mirrored.
14546 In mirrored mode, higher values will be represented on the left
14547 side for @code{row} mode and at the top for @code{column} mode. Default is
14548 @code{1} (mirrored).
14552 It accepts the following values:
14555 Presents information identical to that in the @code{parade}, except
14556 that the graphs representing color components are superimposed directly
14559 This display mode makes it easier to spot relative differences or similarities
14560 in overlapping areas of the color components that are supposed to be identical,
14561 such as neutral whites, grays, or blacks.
14564 Display separate graph for the color components side by side in
14565 @code{row} mode or one below the other in @code{column} mode.
14568 Display separate graph for the color components side by side in
14569 @code{column} mode or one below the other in @code{row} mode.
14571 Using this display mode makes it easy to spot color casts in the highlights
14572 and shadows of an image, by comparing the contours of the top and the bottom
14573 graphs of each waveform. Since whites, grays, and blacks are characterized
14574 by exactly equal amounts of red, green, and blue, neutral areas of the picture
14575 should display three waveforms of roughly equal width/height. If not, the
14576 correction is easy to perform by making level adjustments the three waveforms.
14578 Default is @code{stack}.
14580 @item components, c
14581 Set which color components to display. Default is 1, which means only luminance
14582 or red color component if input is in RGB colorspace. If is set for example to
14583 7 it will display all 3 (if) available color components.
14588 No envelope, this is default.
14591 Instant envelope, minimum and maximum values presented in graph will be easily
14592 visible even with small @code{step} value.
14595 Hold minimum and maximum values presented in graph across time. This way you
14596 can still spot out of range values without constantly looking at waveforms.
14599 Peak and instant envelope combined together.
14605 No filtering, this is default.
14608 Luma and chroma combined together.
14611 Similar as above, but shows difference between blue and red chroma.
14614 Displays only chroma.
14617 Displays actual color value on waveform.
14620 Similar as above, but with luma showing frequency of chroma values.
14624 Set which graticule to display.
14628 Do not display graticule.
14631 Display green graticule showing legal broadcast ranges.
14635 Set graticule opacity.
14638 Set graticule flags.
14642 Draw numbers above lines. By default enabled.
14645 Draw dots instead of lines.
14649 Set scale used for displaying graticule.
14656 Default is digital.
14659 Set background opacity.
14662 @section weave, doubleweave
14664 The @code{weave} takes a field-based video input and join
14665 each two sequential fields into single frame, producing a new double
14666 height clip with half the frame rate and half the frame count.
14668 The @code{doubleweave} works same as @code{weave} but without
14669 halving frame rate and frame count.
14671 It accepts the following option:
14675 Set first field. Available values are:
14679 Set the frame as top-field-first.
14682 Set the frame as bottom-field-first.
14686 @subsection Examples
14690 Interlace video using @ref{select} and @ref{separatefields} filter:
14692 separatefields,select=eq(mod(n,4),0)+eq(mod(n,4),3),weave
14697 Apply the xBR high-quality magnification filter which is designed for pixel
14698 art. It follows a set of edge-detection rules, see
14699 @url{http://www.libretro.com/forums/viewtopic.php?f=6&t=134}.
14701 It accepts the following option:
14705 Set the scaling dimension: @code{2} for @code{2xBR}, @code{3} for
14706 @code{3xBR} and @code{4} for @code{4xBR}.
14707 Default is @code{3}.
14713 Deinterlace the input video ("yadif" means "yet another deinterlacing
14716 It accepts the following parameters:
14722 The interlacing mode to adopt. It accepts one of the following values:
14725 @item 0, send_frame
14726 Output one frame for each frame.
14727 @item 1, send_field
14728 Output one frame for each field.
14729 @item 2, send_frame_nospatial
14730 Like @code{send_frame}, but it skips the spatial interlacing check.
14731 @item 3, send_field_nospatial
14732 Like @code{send_field}, but it skips the spatial interlacing check.
14735 The default value is @code{send_frame}.
14738 The picture field parity assumed for the input interlaced video. It accepts one
14739 of the following values:
14743 Assume the top field is first.
14745 Assume the bottom field is first.
14747 Enable automatic detection of field parity.
14750 The default value is @code{auto}.
14751 If the interlacing is unknown or the decoder does not export this information,
14752 top field first will be assumed.
14755 Specify which frames to deinterlace. Accept one of the following
14760 Deinterlace all frames.
14761 @item 1, interlaced
14762 Only deinterlace frames marked as interlaced.
14765 The default value is @code{all}.
14770 Apply Zoom & Pan effect.
14772 This filter accepts the following options:
14776 Set the zoom expression. Default is 1.
14780 Set the x and y expression. Default is 0.
14783 Set the duration expression in number of frames.
14784 This sets for how many number of frames effect will last for
14785 single input image.
14788 Set the output image size, default is 'hd720'.
14791 Set the output frame rate, default is '25'.
14794 Each expression can contain the following constants:
14813 Output frame count.
14817 Last calculated 'x' and 'y' position from 'x' and 'y' expression
14818 for current input frame.
14822 'x' and 'y' of last output frame of previous input frame or 0 when there was
14823 not yet such frame (first input frame).
14826 Last calculated zoom from 'z' expression for current input frame.
14829 Last calculated zoom of last output frame of previous input frame.
14832 Number of output frames for current input frame. Calculated from 'd' expression
14833 for each input frame.
14836 number of output frames created for previous input frame
14839 Rational number: input width / input height
14842 sample aspect ratio
14845 display aspect ratio
14849 @subsection Examples
14853 Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
14855 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
14859 Zoom-in up to 1.5 and pan always at center of picture:
14861 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
14865 Same as above but without pausing:
14867 zoompan=z='min(max(zoom,pzoom)+0.0015,1.5)':d=1:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
14872 Scale (resize) the input video, using the z.lib library:
14873 https://github.com/sekrit-twc/zimg.
14875 The zscale filter forces the output display aspect ratio to be the same
14876 as the input, by changing the output sample aspect ratio.
14878 If the input image format is different from the format requested by
14879 the next filter, the zscale filter will convert the input to the
14882 @subsection Options
14883 The filter accepts the following options.
14888 Set the output video dimension expression. Default value is the input
14891 If the @var{width} or @var{w} is 0, the input width is used for the output.
14892 If the @var{height} or @var{h} is 0, the input height is used for the output.
14894 If one of the values is -1, the zscale filter will use a value that
14895 maintains the aspect ratio of the input image, calculated from the
14896 other specified dimension. If both of them are -1, the input size is
14899 If one of the values is -n with n > 1, the zscale filter will also use a value
14900 that maintains the aspect ratio of the input image, calculated from the other
14901 specified dimension. After that it will, however, make sure that the calculated
14902 dimension is divisible by n and adjust the value if necessary.
14904 See below for the list of accepted constants for use in the dimension
14908 Set the video size. For the syntax of this option, check the
14909 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14912 Set the dither type.
14914 Possible values are:
14919 @item error_diffusion
14925 Set the resize filter type.
14927 Possible values are:
14937 Default is bilinear.
14940 Set the color range.
14942 Possible values are:
14949 Default is same as input.
14952 Set the color primaries.
14954 Possible values are:
14964 Default is same as input.
14967 Set the transfer characteristics.
14969 Possible values are:
14983 Default is same as input.
14986 Set the colorspace matrix.
14988 Possible value are:
14999 Default is same as input.
15002 Set the input color range.
15004 Possible values are:
15011 Default is same as input.
15013 @item primariesin, pin
15014 Set the input color primaries.
15016 Possible values are:
15026 Default is same as input.
15028 @item transferin, tin
15029 Set the input transfer characteristics.
15031 Possible values are:
15042 Default is same as input.
15044 @item matrixin, min
15045 Set the input colorspace matrix.
15047 Possible value are:
15059 Set the output chroma location.
15061 Possible values are:
15072 @item chromalin, cin
15073 Set the input chroma location.
15075 Possible values are:
15087 Set the nominal peak luminance.
15090 The values of the @option{w} and @option{h} options are expressions
15091 containing the following constants:
15096 The input width and height
15100 These are the same as @var{in_w} and @var{in_h}.
15104 The output (scaled) width and height
15108 These are the same as @var{out_w} and @var{out_h}
15111 The same as @var{iw} / @var{ih}
15114 input sample aspect ratio
15117 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
15121 horizontal and vertical input chroma subsample values. For example for the
15122 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
15126 horizontal and vertical output chroma subsample values. For example for the
15127 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
15133 @c man end VIDEO FILTERS
15135 @chapter Video Sources
15136 @c man begin VIDEO SOURCES
15138 Below is a description of the currently available video sources.
15142 Buffer video frames, and make them available to the filter chain.
15144 This source is mainly intended for a programmatic use, in particular
15145 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
15147 It accepts the following parameters:
15152 Specify the size (width and height) of the buffered video frames. For the
15153 syntax of this option, check the
15154 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15157 The input video width.
15160 The input video height.
15163 A string representing the pixel format of the buffered video frames.
15164 It may be a number corresponding to a pixel format, or a pixel format
15168 Specify the timebase assumed by the timestamps of the buffered frames.
15171 Specify the frame rate expected for the video stream.
15173 @item pixel_aspect, sar
15174 The sample (pixel) aspect ratio of the input video.
15177 Specify the optional parameters to be used for the scale filter which
15178 is automatically inserted when an input change is detected in the
15179 input size or format.
15181 @item hw_frames_ctx
15182 When using a hardware pixel format, this should be a reference to an
15183 AVHWFramesContext describing input frames.
15188 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
15191 will instruct the source to accept video frames with size 320x240 and
15192 with format "yuv410p", assuming 1/24 as the timestamps timebase and
15193 square pixels (1:1 sample aspect ratio).
15194 Since the pixel format with name "yuv410p" corresponds to the number 6
15195 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
15196 this example corresponds to:
15198 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
15201 Alternatively, the options can be specified as a flat string, but this
15202 syntax is deprecated:
15204 @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}]
15208 Create a pattern generated by an elementary cellular automaton.
15210 The initial state of the cellular automaton can be defined through the
15211 @option{filename} and @option{pattern} options. If such options are
15212 not specified an initial state is created randomly.
15214 At each new frame a new row in the video is filled with the result of
15215 the cellular automaton next generation. The behavior when the whole
15216 frame is filled is defined by the @option{scroll} option.
15218 This source accepts the following options:
15222 Read the initial cellular automaton state, i.e. the starting row, from
15223 the specified file.
15224 In the file, each non-whitespace character is considered an alive
15225 cell, a newline will terminate the row, and further characters in the
15226 file will be ignored.
15229 Read the initial cellular automaton state, i.e. the starting row, from
15230 the specified string.
15232 Each non-whitespace character in the string is considered an alive
15233 cell, a newline will terminate the row, and further characters in the
15234 string will be ignored.
15237 Set the video rate, that is the number of frames generated per second.
15240 @item random_fill_ratio, ratio
15241 Set the random fill ratio for the initial cellular automaton row. It
15242 is a floating point number value ranging from 0 to 1, defaults to
15245 This option is ignored when a file or a pattern is specified.
15247 @item random_seed, seed
15248 Set the seed for filling randomly the initial row, must be an integer
15249 included between 0 and UINT32_MAX. If not specified, or if explicitly
15250 set to -1, the filter will try to use a good random seed on a best
15254 Set the cellular automaton rule, it is a number ranging from 0 to 255.
15255 Default value is 110.
15258 Set the size of the output video. For the syntax of this option, check the
15259 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15261 If @option{filename} or @option{pattern} is specified, the size is set
15262 by default to the width of the specified initial state row, and the
15263 height is set to @var{width} * PHI.
15265 If @option{size} is set, it must contain the width of the specified
15266 pattern string, and the specified pattern will be centered in the
15269 If a filename or a pattern string is not specified, the size value
15270 defaults to "320x518" (used for a randomly generated initial state).
15273 If set to 1, scroll the output upward when all the rows in the output
15274 have been already filled. If set to 0, the new generated row will be
15275 written over the top row just after the bottom row is filled.
15278 @item start_full, full
15279 If set to 1, completely fill the output with generated rows before
15280 outputting the first frame.
15281 This is the default behavior, for disabling set the value to 0.
15284 If set to 1, stitch the left and right row edges together.
15285 This is the default behavior, for disabling set the value to 0.
15288 @subsection Examples
15292 Read the initial state from @file{pattern}, and specify an output of
15295 cellauto=f=pattern:s=200x400
15299 Generate a random initial row with a width of 200 cells, with a fill
15302 cellauto=ratio=2/3:s=200x200
15306 Create a pattern generated by rule 18 starting by a single alive cell
15307 centered on an initial row with width 100:
15309 cellauto=p=@@:s=100x400:full=0:rule=18
15313 Specify a more elaborated initial pattern:
15315 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
15320 @anchor{coreimagesrc}
15321 @section coreimagesrc
15322 Video source generated on GPU using Apple's CoreImage API on OSX.
15324 This video source is a specialized version of the @ref{coreimage} video filter.
15325 Use a core image generator at the beginning of the applied filterchain to
15326 generate the content.
15328 The coreimagesrc video source accepts the following options:
15330 @item list_generators
15331 List all available generators along with all their respective options as well as
15332 possible minimum and maximum values along with the default values.
15334 list_generators=true
15338 Specify the size of the sourced video. For the syntax of this option, check the
15339 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15340 The default value is @code{320x240}.
15343 Specify the frame rate of the sourced video, as the number of frames
15344 generated per second. It has to be a string in the format
15345 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15346 number or a valid video frame rate abbreviation. The default value is
15350 Set the sample aspect ratio of the sourced video.
15353 Set the duration of the sourced video. See
15354 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15355 for the accepted syntax.
15357 If not specified, or the expressed duration is negative, the video is
15358 supposed to be generated forever.
15361 Additionally, all options of the @ref{coreimage} video filter are accepted.
15362 A complete filterchain can be used for further processing of the
15363 generated input without CPU-HOST transfer. See @ref{coreimage} documentation
15364 and examples for details.
15366 @subsection Examples
15371 Use CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
15372 given as complete and escaped command-line for Apple's standard bash shell:
15374 ffmpeg -f lavfi -i coreimagesrc=s=100x100:filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
15376 This example is equivalent to the QRCode example of @ref{coreimage} without the
15377 need for a nullsrc video source.
15381 @section mandelbrot
15383 Generate a Mandelbrot set fractal, and progressively zoom towards the
15384 point specified with @var{start_x} and @var{start_y}.
15386 This source accepts the following options:
15391 Set the terminal pts value. Default value is 400.
15394 Set the terminal scale value.
15395 Must be a floating point value. Default value is 0.3.
15398 Set the inner coloring mode, that is the algorithm used to draw the
15399 Mandelbrot fractal internal region.
15401 It shall assume one of the following values:
15406 Show time until convergence.
15408 Set color based on point closest to the origin of the iterations.
15413 Default value is @var{mincol}.
15416 Set the bailout value. Default value is 10.0.
15419 Set the maximum of iterations performed by the rendering
15420 algorithm. Default value is 7189.
15423 Set outer coloring mode.
15424 It shall assume one of following values:
15426 @item iteration_count
15427 Set iteration cound mode.
15428 @item normalized_iteration_count
15429 set normalized iteration count mode.
15431 Default value is @var{normalized_iteration_count}.
15434 Set frame rate, expressed as number of frames per second. Default
15438 Set frame size. For the syntax of this option, check the "Video
15439 size" section in the ffmpeg-utils manual. Default value is "640x480".
15442 Set the initial scale value. Default value is 3.0.
15445 Set the initial x position. Must be a floating point value between
15446 -100 and 100. Default value is -0.743643887037158704752191506114774.
15449 Set the initial y position. Must be a floating point value between
15450 -100 and 100. Default value is -0.131825904205311970493132056385139.
15455 Generate various test patterns, as generated by the MPlayer test filter.
15457 The size of the generated video is fixed, and is 256x256.
15458 This source is useful in particular for testing encoding features.
15460 This source accepts the following options:
15465 Specify the frame rate of the sourced video, as the number of frames
15466 generated per second. It has to be a string in the format
15467 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15468 number or a valid video frame rate abbreviation. The default value is
15472 Set the duration of the sourced video. See
15473 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15474 for the accepted syntax.
15476 If not specified, or the expressed duration is negative, the video is
15477 supposed to be generated forever.
15481 Set the number or the name of the test to perform. Supported tests are:
15497 Default value is "all", which will cycle through the list of all tests.
15502 mptestsrc=t=dc_luma
15505 will generate a "dc_luma" test pattern.
15507 @section frei0r_src
15509 Provide a frei0r source.
15511 To enable compilation of this filter you need to install the frei0r
15512 header and configure FFmpeg with @code{--enable-frei0r}.
15514 This source accepts the following parameters:
15519 The size of the video to generate. For the syntax of this option, check the
15520 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15523 The framerate of the generated video. It may be a string of the form
15524 @var{num}/@var{den} or a frame rate abbreviation.
15527 The name to the frei0r source to load. For more information regarding frei0r and
15528 how to set the parameters, read the @ref{frei0r} section in the video filters
15531 @item filter_params
15532 A '|'-separated list of parameters to pass to the frei0r source.
15536 For example, to generate a frei0r partik0l source with size 200x200
15537 and frame rate 10 which is overlaid on the overlay filter main input:
15539 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
15544 Generate a life pattern.
15546 This source is based on a generalization of John Conway's life game.
15548 The sourced input represents a life grid, each pixel represents a cell
15549 which can be in one of two possible states, alive or dead. Every cell
15550 interacts with its eight neighbours, which are the cells that are
15551 horizontally, vertically, or diagonally adjacent.
15553 At each interaction the grid evolves according to the adopted rule,
15554 which specifies the number of neighbor alive cells which will make a
15555 cell stay alive or born. The @option{rule} option allows one to specify
15558 This source accepts the following options:
15562 Set the file from which to read the initial grid state. In the file,
15563 each non-whitespace character is considered an alive cell, and newline
15564 is used to delimit the end of each row.
15566 If this option is not specified, the initial grid is generated
15570 Set the video rate, that is the number of frames generated per second.
15573 @item random_fill_ratio, ratio
15574 Set the random fill ratio for the initial random grid. It is a
15575 floating point number value ranging from 0 to 1, defaults to 1/PHI.
15576 It is ignored when a file is specified.
15578 @item random_seed, seed
15579 Set the seed for filling the initial random grid, must be an integer
15580 included between 0 and UINT32_MAX. If not specified, or if explicitly
15581 set to -1, the filter will try to use a good random seed on a best
15587 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
15588 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
15589 @var{NS} specifies the number of alive neighbor cells which make a
15590 live cell stay alive, and @var{NB} the number of alive neighbor cells
15591 which make a dead cell to become alive (i.e. to "born").
15592 "s" and "b" can be used in place of "S" and "B", respectively.
15594 Alternatively a rule can be specified by an 18-bits integer. The 9
15595 high order bits are used to encode the next cell state if it is alive
15596 for each number of neighbor alive cells, the low order bits specify
15597 the rule for "borning" new cells. Higher order bits encode for an
15598 higher number of neighbor cells.
15599 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
15600 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
15602 Default value is "S23/B3", which is the original Conway's game of life
15603 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
15604 cells, and will born a new cell if there are three alive cells around
15608 Set the size of the output video. For the syntax of this option, check the
15609 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15611 If @option{filename} is specified, the size is set by default to the
15612 same size of the input file. If @option{size} is set, it must contain
15613 the size specified in the input file, and the initial grid defined in
15614 that file is centered in the larger resulting area.
15616 If a filename is not specified, the size value defaults to "320x240"
15617 (used for a randomly generated initial grid).
15620 If set to 1, stitch the left and right grid edges together, and the
15621 top and bottom edges also. Defaults to 1.
15624 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
15625 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
15626 value from 0 to 255.
15629 Set the color of living (or new born) cells.
15632 Set the color of dead cells. If @option{mold} is set, this is the first color
15633 used to represent a dead cell.
15636 Set mold color, for definitely dead and moldy cells.
15638 For the syntax of these 3 color options, check the "Color" section in the
15639 ffmpeg-utils manual.
15642 @subsection Examples
15646 Read a grid from @file{pattern}, and center it on a grid of size
15649 life=f=pattern:s=300x300
15653 Generate a random grid of size 200x200, with a fill ratio of 2/3:
15655 life=ratio=2/3:s=200x200
15659 Specify a custom rule for evolving a randomly generated grid:
15665 Full example with slow death effect (mold) using @command{ffplay}:
15667 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
15674 @anchor{haldclutsrc}
15676 @anchor{rgbtestsrc}
15678 @anchor{smptehdbars}
15681 @anchor{yuvtestsrc}
15682 @section allrgb, allyuv, color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc, testsrc2, yuvtestsrc
15684 The @code{allrgb} source returns frames of size 4096x4096 of all rgb colors.
15686 The @code{allyuv} source returns frames of size 4096x4096 of all yuv colors.
15688 The @code{color} source provides an uniformly colored input.
15690 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
15691 @ref{haldclut} filter.
15693 The @code{nullsrc} source returns unprocessed video frames. It is
15694 mainly useful to be employed in analysis / debugging tools, or as the
15695 source for filters which ignore the input data.
15697 The @code{rgbtestsrc} source generates an RGB test pattern useful for
15698 detecting RGB vs BGR issues. You should see a red, green and blue
15699 stripe from top to bottom.
15701 The @code{smptebars} source generates a color bars pattern, based on
15702 the SMPTE Engineering Guideline EG 1-1990.
15704 The @code{smptehdbars} source generates a color bars pattern, based on
15705 the SMPTE RP 219-2002.
15707 The @code{testsrc} source generates a test video pattern, showing a
15708 color pattern, a scrolling gradient and a timestamp. This is mainly
15709 intended for testing purposes.
15711 The @code{testsrc2} source is similar to testsrc, but supports more
15712 pixel formats instead of just @code{rgb24}. This allows using it as an
15713 input for other tests without requiring a format conversion.
15715 The @code{yuvtestsrc} source generates an YUV test pattern. You should
15716 see a y, cb and cr stripe from top to bottom.
15718 The sources accept the following parameters:
15723 Specify the color of the source, only available in the @code{color}
15724 source. For the syntax of this option, check the "Color" section in the
15725 ffmpeg-utils manual.
15728 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
15729 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
15730 pixels to be used as identity matrix for 3D lookup tables. Each component is
15731 coded on a @code{1/(N*N)} scale.
15734 Specify the size of the sourced video. For the syntax of this option, check the
15735 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15736 The default value is @code{320x240}.
15738 This option is not available with the @code{haldclutsrc} filter.
15741 Specify the frame rate of the sourced video, as the number of frames
15742 generated per second. It has to be a string in the format
15743 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
15744 number or a valid video frame rate abbreviation. The default value is
15748 Set the sample aspect ratio of the sourced video.
15751 Set the duration of the sourced video. See
15752 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
15753 for the accepted syntax.
15755 If not specified, or the expressed duration is negative, the video is
15756 supposed to be generated forever.
15759 Set the number of decimals to show in the timestamp, only available in the
15760 @code{testsrc} source.
15762 The displayed timestamp value will correspond to the original
15763 timestamp value multiplied by the power of 10 of the specified
15764 value. Default value is 0.
15767 For example the following:
15769 testsrc=duration=5.3:size=qcif:rate=10
15772 will generate a video with a duration of 5.3 seconds, with size
15773 176x144 and a frame rate of 10 frames per second.
15775 The following graph description will generate a red source
15776 with an opacity of 0.2, with size "qcif" and a frame rate of 10
15779 color=c=red@@0.2:s=qcif:r=10
15782 If the input content is to be ignored, @code{nullsrc} can be used. The
15783 following command generates noise in the luminance plane by employing
15784 the @code{geq} filter:
15786 nullsrc=s=256x256, geq=random(1)*255:128:128
15789 @subsection Commands
15791 The @code{color} source supports the following commands:
15795 Set the color of the created image. Accepts the same syntax of the
15796 corresponding @option{color} option.
15799 @c man end VIDEO SOURCES
15801 @chapter Video Sinks
15802 @c man begin VIDEO SINKS
15804 Below is a description of the currently available video sinks.
15806 @section buffersink
15808 Buffer video frames, and make them available to the end of the filter
15811 This sink is mainly intended for programmatic use, in particular
15812 through the interface defined in @file{libavfilter/buffersink.h}
15813 or the options system.
15815 It accepts a pointer to an AVBufferSinkContext structure, which
15816 defines the incoming buffers' formats, to be passed as the opaque
15817 parameter to @code{avfilter_init_filter} for initialization.
15821 Null video sink: do absolutely nothing with the input video. It is
15822 mainly useful as a template and for use in analysis / debugging
15825 @c man end VIDEO SINKS
15827 @chapter Multimedia Filters
15828 @c man begin MULTIMEDIA FILTERS
15830 Below is a description of the currently available multimedia filters.
15834 Convert input audio to a video output, displaying the audio bit scope.
15836 The filter accepts the following options:
15840 Set frame rate, expressed as number of frames per second. Default
15844 Specify the video size for the output. For the syntax of this option, check the
15845 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15846 Default value is @code{1024x256}.
15849 Specify list of colors separated by space or by '|' which will be used to
15850 draw channels. Unrecognized or missing colors will be replaced
15854 @section ahistogram
15856 Convert input audio to a video output, displaying the volume histogram.
15858 The filter accepts the following options:
15862 Specify how histogram is calculated.
15864 It accepts the following values:
15867 Use single histogram for all channels.
15869 Use separate histogram for each channel.
15871 Default is @code{single}.
15874 Set frame rate, expressed as number of frames per second. Default
15878 Specify the video size for the output. For the syntax of this option, check the
15879 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15880 Default value is @code{hd720}.
15885 It accepts the following values:
15896 reverse logarithmic
15898 Default is @code{log}.
15901 Set amplitude scale.
15903 It accepts the following values:
15910 Default is @code{log}.
15913 Set how much frames to accumulate in histogram.
15914 Defauls is 1. Setting this to -1 accumulates all frames.
15917 Set histogram ratio of window height.
15920 Set sonogram sliding.
15922 It accepts the following values:
15925 replace old rows with new ones.
15927 scroll from top to bottom.
15929 Default is @code{replace}.
15932 @section aphasemeter
15934 Convert input audio to a video output, displaying the audio phase.
15936 The filter accepts the following options:
15940 Set the output frame rate. Default value is @code{25}.
15943 Set the video size for the output. For the syntax of this option, check the
15944 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15945 Default value is @code{800x400}.
15950 Specify the red, green, blue contrast. Default values are @code{2},
15951 @code{7} and @code{1}.
15952 Allowed range is @code{[0, 255]}.
15955 Set color which will be used for drawing median phase. If color is
15956 @code{none} which is default, no median phase value will be drawn.
15959 Enable video output. Default is enabled.
15962 The filter also exports the frame metadata @code{lavfi.aphasemeter.phase} which
15963 represents mean phase of current audio frame. Value is in range @code{[-1, 1]}.
15964 The @code{-1} means left and right channels are completely out of phase and
15965 @code{1} means channels are in phase.
15967 @section avectorscope
15969 Convert input audio to a video output, representing the audio vector
15972 The filter is used to measure the difference between channels of stereo
15973 audio stream. A monoaural signal, consisting of identical left and right
15974 signal, results in straight vertical line. Any stereo separation is visible
15975 as a deviation from this line, creating a Lissajous figure.
15976 If the straight (or deviation from it) but horizontal line appears this
15977 indicates that the left and right channels are out of phase.
15979 The filter accepts the following options:
15983 Set the vectorscope mode.
15985 Available values are:
15988 Lissajous rotated by 45 degrees.
15991 Same as above but not rotated.
15994 Shape resembling half of circle.
15997 Default value is @samp{lissajous}.
16000 Set the video size for the output. For the syntax of this option, check the
16001 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16002 Default value is @code{400x400}.
16005 Set the output frame rate. Default value is @code{25}.
16011 Specify the red, green, blue and alpha contrast. Default values are @code{40},
16012 @code{160}, @code{80} and @code{255}.
16013 Allowed range is @code{[0, 255]}.
16019 Specify the red, green, blue and alpha fade. Default values are @code{15},
16020 @code{10}, @code{5} and @code{5}.
16021 Allowed range is @code{[0, 255]}.
16024 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
16027 Set the vectorscope drawing mode.
16029 Available values are:
16032 Draw dot for each sample.
16035 Draw line between previous and current sample.
16038 Default value is @samp{dot}.
16041 Specify amplitude scale of audio samples.
16043 Available values are:
16060 @subsection Examples
16064 Complete example using @command{ffplay}:
16066 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
16067 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
16071 @section bench, abench
16073 Benchmark part of a filtergraph.
16075 The filter accepts the following options:
16079 Start or stop a timer.
16081 Available values are:
16084 Get the current time, set it as frame metadata (using the key
16085 @code{lavfi.bench.start_time}), and forward the frame to the next filter.
16088 Get the current time and fetch the @code{lavfi.bench.start_time} metadata from
16089 the input frame metadata to get the time difference. Time difference, average,
16090 maximum and minimum time (respectively @code{t}, @code{avg}, @code{max} and
16091 @code{min}) are then printed. The timestamps are expressed in seconds.
16095 @subsection Examples
16099 Benchmark @ref{selectivecolor} filter:
16101 bench=start,selectivecolor=reds=-.2 .12 -.49,bench=stop
16107 Concatenate audio and video streams, joining them together one after the
16110 The filter works on segments of synchronized video and audio streams. All
16111 segments must have the same number of streams of each type, and that will
16112 also be the number of streams at output.
16114 The filter accepts the following options:
16119 Set the number of segments. Default is 2.
16122 Set the number of output video streams, that is also the number of video
16123 streams in each segment. Default is 1.
16126 Set the number of output audio streams, that is also the number of audio
16127 streams in each segment. Default is 0.
16130 Activate unsafe mode: do not fail if segments have a different format.
16134 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
16135 @var{a} audio outputs.
16137 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
16138 segment, in the same order as the outputs, then the inputs for the second
16141 Related streams do not always have exactly the same duration, for various
16142 reasons including codec frame size or sloppy authoring. For that reason,
16143 related synchronized streams (e.g. a video and its audio track) should be
16144 concatenated at once. The concat filter will use the duration of the longest
16145 stream in each segment (except the last one), and if necessary pad shorter
16146 audio streams with silence.
16148 For this filter to work correctly, all segments must start at timestamp 0.
16150 All corresponding streams must have the same parameters in all segments; the
16151 filtering system will automatically select a common pixel format for video
16152 streams, and a common sample format, sample rate and channel layout for
16153 audio streams, but other settings, such as resolution, must be converted
16154 explicitly by the user.
16156 Different frame rates are acceptable but will result in variable frame rate
16157 at output; be sure to configure the output file to handle it.
16159 @subsection Examples
16163 Concatenate an opening, an episode and an ending, all in bilingual version
16164 (video in stream 0, audio in streams 1 and 2):
16166 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
16167 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
16168 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
16169 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
16173 Concatenate two parts, handling audio and video separately, using the
16174 (a)movie sources, and adjusting the resolution:
16176 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
16177 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
16178 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
16180 Note that a desync will happen at the stitch if the audio and video streams
16181 do not have exactly the same duration in the first file.
16185 @section drawgraph, adrawgraph
16187 Draw a graph using input video or audio metadata.
16189 It accepts the following parameters:
16193 Set 1st frame metadata key from which metadata values will be used to draw a graph.
16196 Set 1st foreground color expression.
16199 Set 2nd frame metadata key from which metadata values will be used to draw a graph.
16202 Set 2nd foreground color expression.
16205 Set 3rd frame metadata key from which metadata values will be used to draw a graph.
16208 Set 3rd foreground color expression.
16211 Set 4th frame metadata key from which metadata values will be used to draw a graph.
16214 Set 4th foreground color expression.
16217 Set minimal value of metadata value.
16220 Set maximal value of metadata value.
16223 Set graph background color. Default is white.
16228 Available values for mode is:
16235 Default is @code{line}.
16240 Available values for slide is:
16243 Draw new frame when right border is reached.
16246 Replace old columns with new ones.
16249 Scroll from right to left.
16252 Scroll from left to right.
16255 Draw single picture.
16258 Default is @code{frame}.
16261 Set size of graph video. For the syntax of this option, check the
16262 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16263 The default value is @code{900x256}.
16265 The foreground color expressions can use the following variables:
16268 Minimal value of metadata value.
16271 Maximal value of metadata value.
16274 Current metadata key value.
16277 The color is defined as 0xAABBGGRR.
16280 Example using metadata from @ref{signalstats} filter:
16282 signalstats,drawgraph=lavfi.signalstats.YAVG:min=0:max=255
16285 Example using metadata from @ref{ebur128} filter:
16287 ebur128=metadata=1,adrawgraph=lavfi.r128.M:min=-120:max=5
16293 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
16294 it unchanged. By default, it logs a message at a frequency of 10Hz with the
16295 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
16296 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
16298 The filter also has a video output (see the @var{video} option) with a real
16299 time graph to observe the loudness evolution. The graphic contains the logged
16300 message mentioned above, so it is not printed anymore when this option is set,
16301 unless the verbose logging is set. The main graphing area contains the
16302 short-term loudness (3 seconds of analysis), and the gauge on the right is for
16303 the momentary loudness (400 milliseconds).
16305 More information about the Loudness Recommendation EBU R128 on
16306 @url{http://tech.ebu.ch/loudness}.
16308 The filter accepts the following options:
16313 Activate the video output. The audio stream is passed unchanged whether this
16314 option is set or no. The video stream will be the first output stream if
16315 activated. Default is @code{0}.
16318 Set the video size. This option is for video only. For the syntax of this
16320 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16321 Default and minimum resolution is @code{640x480}.
16324 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
16325 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
16326 other integer value between this range is allowed.
16329 Set metadata injection. If set to @code{1}, the audio input will be segmented
16330 into 100ms output frames, each of them containing various loudness information
16331 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
16333 Default is @code{0}.
16336 Force the frame logging level.
16338 Available values are:
16341 information logging level
16343 verbose logging level
16346 By default, the logging level is set to @var{info}. If the @option{video} or
16347 the @option{metadata} options are set, it switches to @var{verbose}.
16352 Available modes can be cumulated (the option is a @code{flag} type). Possible
16356 Disable any peak mode (default).
16358 Enable sample-peak mode.
16360 Simple peak mode looking for the higher sample value. It logs a message
16361 for sample-peak (identified by @code{SPK}).
16363 Enable true-peak mode.
16365 If enabled, the peak lookup is done on an over-sampled version of the input
16366 stream for better peak accuracy. It logs a message for true-peak.
16367 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
16368 This mode requires a build with @code{libswresample}.
16372 Treat mono input files as "dual mono". If a mono file is intended for playback
16373 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
16374 If set to @code{true}, this option will compensate for this effect.
16375 Multi-channel input files are not affected by this option.
16378 Set a specific pan law to be used for the measurement of dual mono files.
16379 This parameter is optional, and has a default value of -3.01dB.
16382 @subsection Examples
16386 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
16388 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
16392 Run an analysis with @command{ffmpeg}:
16394 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
16398 @section interleave, ainterleave
16400 Temporally interleave frames from several inputs.
16402 @code{interleave} works with video inputs, @code{ainterleave} with audio.
16404 These filters read frames from several inputs and send the oldest
16405 queued frame to the output.
16407 Input streams must have well defined, monotonically increasing frame
16410 In order to submit one frame to output, these filters need to enqueue
16411 at least one frame for each input, so they cannot work in case one
16412 input is not yet terminated and will not receive incoming frames.
16414 For example consider the case when one input is a @code{select} filter
16415 which always drops input frames. The @code{interleave} filter will keep
16416 reading from that input, but it will never be able to send new frames
16417 to output until the input sends an end-of-stream signal.
16419 Also, depending on inputs synchronization, the filters will drop
16420 frames in case one input receives more frames than the other ones, and
16421 the queue is already filled.
16423 These filters accept the following options:
16427 Set the number of different inputs, it is 2 by default.
16430 @subsection Examples
16434 Interleave frames belonging to different streams using @command{ffmpeg}:
16436 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
16440 Add flickering blur effect:
16442 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
16446 @section metadata, ametadata
16448 Manipulate frame metadata.
16450 This filter accepts the following options:
16454 Set mode of operation of the filter.
16456 Can be one of the following:
16460 If both @code{value} and @code{key} is set, select frames
16461 which have such metadata. If only @code{key} is set, select
16462 every frame that has such key in metadata.
16465 Add new metadata @code{key} and @code{value}. If key is already available
16469 Modify value of already present key.
16472 If @code{value} is set, delete only keys that have such value.
16473 Otherwise, delete key. If @code{key} is not set, delete all metadata values in
16477 Print key and its value if metadata was found. If @code{key} is not set print all
16478 metadata values available in frame.
16482 Set key used with all modes. Must be set for all modes except @code{print} and @code{delete}.
16485 Set metadata value which will be used. This option is mandatory for
16486 @code{modify} and @code{add} mode.
16489 Which function to use when comparing metadata value and @code{value}.
16491 Can be one of following:
16495 Values are interpreted as strings, returns true if metadata value is same as @code{value}.
16498 Values are interpreted as strings, returns true if metadata value starts with
16499 the @code{value} option string.
16502 Values are interpreted as floats, returns true if metadata value is less than @code{value}.
16505 Values are interpreted as floats, returns true if @code{value} is equal with metadata value.
16508 Values are interpreted as floats, returns true if metadata value is greater than @code{value}.
16511 Values are interpreted as floats, returns true if expression from option @code{expr}
16516 Set expression which is used when @code{function} is set to @code{expr}.
16517 The expression is evaluated through the eval API and can contain the following
16522 Float representation of @code{value} from metadata key.
16525 Float representation of @code{value} as supplied by user in @code{value} option.
16529 If specified in @code{print} mode, output is written to the named file. Instead of
16530 plain filename any writable url can be specified. Filename ``-'' is a shorthand
16531 for standard output. If @code{file} option is not set, output is written to the log
16532 with AV_LOG_INFO loglevel.
16536 @subsection Examples
16540 Print all metadata values for frames with key @code{lavfi.singnalstats.YDIF} with values
16543 signalstats,metadata=print:key=lavfi.signalstats.YDIF:value=0:function=expr:expr='between(VALUE1,0,1)'
16546 Print silencedetect output to file @file{metadata.txt}.
16548 silencedetect,ametadata=mode=print:file=metadata.txt
16551 Direct all metadata to a pipe with file descriptor 4.
16553 metadata=mode=print:file='pipe\:4'
16557 @section perms, aperms
16559 Set read/write permissions for the output frames.
16561 These filters are mainly aimed at developers to test direct path in the
16562 following filter in the filtergraph.
16564 The filters accept the following options:
16568 Select the permissions mode.
16570 It accepts the following values:
16573 Do nothing. This is the default.
16575 Set all the output frames read-only.
16577 Set all the output frames directly writable.
16579 Make the frame read-only if writable, and writable if read-only.
16581 Set each output frame read-only or writable randomly.
16585 Set the seed for the @var{random} mode, must be an integer included between
16586 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
16587 @code{-1}, the filter will try to use a good random seed on a best effort
16591 Note: in case of auto-inserted filter between the permission filter and the
16592 following one, the permission might not be received as expected in that
16593 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
16594 perms/aperms filter can avoid this problem.
16596 @section realtime, arealtime
16598 Slow down filtering to match real time approximatively.
16600 These filters will pause the filtering for a variable amount of time to
16601 match the output rate with the input timestamps.
16602 They are similar to the @option{re} option to @code{ffmpeg}.
16604 They accept the following options:
16608 Time limit for the pauses. Any pause longer than that will be considered
16609 a timestamp discontinuity and reset the timer. Default is 2 seconds.
16613 @section select, aselect
16615 Select frames to pass in output.
16617 This filter accepts the following options:
16622 Set expression, which is evaluated for each input frame.
16624 If the expression is evaluated to zero, the frame is discarded.
16626 If the evaluation result is negative or NaN, the frame is sent to the
16627 first output; otherwise it is sent to the output with index
16628 @code{ceil(val)-1}, assuming that the input index starts from 0.
16630 For example a value of @code{1.2} corresponds to the output with index
16631 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
16634 Set the number of outputs. The output to which to send the selected
16635 frame is based on the result of the evaluation. Default value is 1.
16638 The expression can contain the following constants:
16642 The (sequential) number of the filtered frame, starting from 0.
16645 The (sequential) number of the selected frame, starting from 0.
16647 @item prev_selected_n
16648 The sequential number of the last selected frame. It's NAN if undefined.
16651 The timebase of the input timestamps.
16654 The PTS (Presentation TimeStamp) of the filtered video frame,
16655 expressed in @var{TB} units. It's NAN if undefined.
16658 The PTS of the filtered video frame,
16659 expressed in seconds. It's NAN if undefined.
16662 The PTS of the previously filtered video frame. It's NAN if undefined.
16664 @item prev_selected_pts
16665 The PTS of the last previously filtered video frame. It's NAN if undefined.
16667 @item prev_selected_t
16668 The PTS of the last previously selected video frame. It's NAN if undefined.
16671 The PTS of the first video frame in the video. It's NAN if undefined.
16674 The time of the first video frame in the video. It's NAN if undefined.
16676 @item pict_type @emph{(video only)}
16677 The type of the filtered frame. It can assume one of the following
16689 @item interlace_type @emph{(video only)}
16690 The frame interlace type. It can assume one of the following values:
16693 The frame is progressive (not interlaced).
16695 The frame is top-field-first.
16697 The frame is bottom-field-first.
16700 @item consumed_sample_n @emph{(audio only)}
16701 the number of selected samples before the current frame
16703 @item samples_n @emph{(audio only)}
16704 the number of samples in the current frame
16706 @item sample_rate @emph{(audio only)}
16707 the input sample rate
16710 This is 1 if the filtered frame is a key-frame, 0 otherwise.
16713 the position in the file of the filtered frame, -1 if the information
16714 is not available (e.g. for synthetic video)
16716 @item scene @emph{(video only)}
16717 value between 0 and 1 to indicate a new scene; a low value reflects a low
16718 probability for the current frame to introduce a new scene, while a higher
16719 value means the current frame is more likely to be one (see the example below)
16721 @item concatdec_select
16722 The concat demuxer can select only part of a concat input file by setting an
16723 inpoint and an outpoint, but the output packets may not be entirely contained
16724 in the selected interval. By using this variable, it is possible to skip frames
16725 generated by the concat demuxer which are not exactly contained in the selected
16728 This works by comparing the frame pts against the @var{lavf.concat.start_time}
16729 and the @var{lavf.concat.duration} packet metadata values which are also
16730 present in the decoded frames.
16732 The @var{concatdec_select} variable is -1 if the frame pts is at least
16733 start_time and either the duration metadata is missing or the frame pts is less
16734 than start_time + duration, 0 otherwise, and NaN if the start_time metadata is
16737 That basically means that an input frame is selected if its pts is within the
16738 interval set by the concat demuxer.
16742 The default value of the select expression is "1".
16744 @subsection Examples
16748 Select all frames in input:
16753 The example above is the same as:
16765 Select only I-frames:
16767 select='eq(pict_type\,I)'
16771 Select one frame every 100:
16773 select='not(mod(n\,100))'
16777 Select only frames contained in the 10-20 time interval:
16779 select=between(t\,10\,20)
16783 Select only I-frames contained in the 10-20 time interval:
16785 select=between(t\,10\,20)*eq(pict_type\,I)
16789 Select frames with a minimum distance of 10 seconds:
16791 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
16795 Use aselect to select only audio frames with samples number > 100:
16797 aselect='gt(samples_n\,100)'
16801 Create a mosaic of the first scenes:
16803 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
16806 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
16810 Send even and odd frames to separate outputs, and compose them:
16812 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
16816 Select useful frames from an ffconcat file which is using inpoints and
16817 outpoints but where the source files are not intra frame only.
16819 ffmpeg -copyts -vsync 0 -segment_time_metadata 1 -i input.ffconcat -vf select=concatdec_select -af aselect=concatdec_select output.avi
16823 @section sendcmd, asendcmd
16825 Send commands to filters in the filtergraph.
16827 These filters read commands to be sent to other filters in the
16830 @code{sendcmd} must be inserted between two video filters,
16831 @code{asendcmd} must be inserted between two audio filters, but apart
16832 from that they act the same way.
16834 The specification of commands can be provided in the filter arguments
16835 with the @var{commands} option, or in a file specified by the
16836 @var{filename} option.
16838 These filters accept the following options:
16841 Set the commands to be read and sent to the other filters.
16843 Set the filename of the commands to be read and sent to the other
16847 @subsection Commands syntax
16849 A commands description consists of a sequence of interval
16850 specifications, comprising a list of commands to be executed when a
16851 particular event related to that interval occurs. The occurring event
16852 is typically the current frame time entering or leaving a given time
16855 An interval is specified by the following syntax:
16857 @var{START}[-@var{END}] @var{COMMANDS};
16860 The time interval is specified by the @var{START} and @var{END} times.
16861 @var{END} is optional and defaults to the maximum time.
16863 The current frame time is considered within the specified interval if
16864 it is included in the interval [@var{START}, @var{END}), that is when
16865 the time is greater or equal to @var{START} and is lesser than
16868 @var{COMMANDS} consists of a sequence of one or more command
16869 specifications, separated by ",", relating to that interval. The
16870 syntax of a command specification is given by:
16872 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
16875 @var{FLAGS} is optional and specifies the type of events relating to
16876 the time interval which enable sending the specified command, and must
16877 be a non-null sequence of identifier flags separated by "+" or "|" and
16878 enclosed between "[" and "]".
16880 The following flags are recognized:
16883 The command is sent when the current frame timestamp enters the
16884 specified interval. In other words, the command is sent when the
16885 previous frame timestamp was not in the given interval, and the
16889 The command is sent when the current frame timestamp leaves the
16890 specified interval. In other words, the command is sent when the
16891 previous frame timestamp was in the given interval, and the
16895 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
16898 @var{TARGET} specifies the target of the command, usually the name of
16899 the filter class or a specific filter instance name.
16901 @var{COMMAND} specifies the name of the command for the target filter.
16903 @var{ARG} is optional and specifies the optional list of argument for
16904 the given @var{COMMAND}.
16906 Between one interval specification and another, whitespaces, or
16907 sequences of characters starting with @code{#} until the end of line,
16908 are ignored and can be used to annotate comments.
16910 A simplified BNF description of the commands specification syntax
16913 @var{COMMAND_FLAG} ::= "enter" | "leave"
16914 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
16915 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
16916 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
16917 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
16918 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
16921 @subsection Examples
16925 Specify audio tempo change at second 4:
16927 asendcmd=c='4.0 atempo tempo 1.5',atempo
16931 Specify a list of drawtext and hue commands in a file.
16933 # show text in the interval 5-10
16934 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
16935 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
16937 # desaturate the image in the interval 15-20
16938 15.0-20.0 [enter] hue s 0,
16939 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
16941 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
16943 # apply an exponential saturation fade-out effect, starting from time 25
16944 25 [enter] hue s exp(25-t)
16947 A filtergraph allowing to read and process the above command list
16948 stored in a file @file{test.cmd}, can be specified with:
16950 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
16955 @section setpts, asetpts
16957 Change the PTS (presentation timestamp) of the input frames.
16959 @code{setpts} works on video frames, @code{asetpts} on audio frames.
16961 This filter accepts the following options:
16966 The expression which is evaluated for each frame to construct its timestamp.
16970 The expression is evaluated through the eval API and can contain the following
16975 frame rate, only defined for constant frame-rate video
16978 The presentation timestamp in input
16981 The count of the input frame for video or the number of consumed samples,
16982 not including the current frame for audio, starting from 0.
16984 @item NB_CONSUMED_SAMPLES
16985 The number of consumed samples, not including the current frame (only
16988 @item NB_SAMPLES, S
16989 The number of samples in the current frame (only audio)
16991 @item SAMPLE_RATE, SR
16992 The audio sample rate.
16995 The PTS of the first frame.
16998 the time in seconds of the first frame
17001 State whether the current frame is interlaced.
17004 the time in seconds of the current frame
17007 original position in the file of the frame, or undefined if undefined
17008 for the current frame
17011 The previous input PTS.
17014 previous input time in seconds
17017 The previous output PTS.
17020 previous output time in seconds
17023 The wallclock (RTC) time in microseconds. This is deprecated, use time(0)
17027 The wallclock (RTC) time at the start of the movie in microseconds.
17030 The timebase of the input timestamps.
17034 @subsection Examples
17038 Start counting PTS from zero
17040 setpts=PTS-STARTPTS
17044 Apply fast motion effect:
17050 Apply slow motion effect:
17056 Set fixed rate of 25 frames per second:
17062 Set fixed rate 25 fps with some jitter:
17064 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
17068 Apply an offset of 10 seconds to the input PTS:
17074 Generate timestamps from a "live source" and rebase onto the current timebase:
17076 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
17080 Generate timestamps by counting samples:
17087 @section settb, asettb
17089 Set the timebase to use for the output frames timestamps.
17090 It is mainly useful for testing timebase configuration.
17092 It accepts the following parameters:
17097 The expression which is evaluated into the output timebase.
17101 The value for @option{tb} is an arithmetic expression representing a
17102 rational. The expression can contain the constants "AVTB" (the default
17103 timebase), "intb" (the input timebase) and "sr" (the sample rate,
17104 audio only). Default value is "intb".
17106 @subsection Examples
17110 Set the timebase to 1/25:
17116 Set the timebase to 1/10:
17122 Set the timebase to 1001/1000:
17128 Set the timebase to 2*intb:
17134 Set the default timebase value:
17141 Convert input audio to a video output representing frequency spectrum
17142 logarithmically using Brown-Puckette constant Q transform algorithm with
17143 direct frequency domain coefficient calculation (but the transform itself
17144 is not really constant Q, instead the Q factor is actually variable/clamped),
17145 with musical tone scale, from E0 to D#10.
17147 The filter accepts the following options:
17151 Specify the video size for the output. It must be even. For the syntax of this option,
17152 check the @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17153 Default value is @code{1920x1080}.
17156 Set the output frame rate. Default value is @code{25}.
17159 Set the bargraph height. It must be even. Default value is @code{-1} which
17160 computes the bargraph height automatically.
17163 Set the axis height. It must be even. Default value is @code{-1} which computes
17164 the axis height automatically.
17167 Set the sonogram height. It must be even. Default value is @code{-1} which
17168 computes the sonogram height automatically.
17171 Set the fullhd resolution. This option is deprecated, use @var{size}, @var{s}
17172 instead. Default value is @code{1}.
17174 @item sono_v, volume
17175 Specify the sonogram volume expression. It can contain variables:
17178 the @var{bar_v} evaluated expression
17179 @item frequency, freq, f
17180 the frequency where it is evaluated
17181 @item timeclamp, tc
17182 the value of @var{timeclamp} option
17186 @item a_weighting(f)
17187 A-weighting of equal loudness
17188 @item b_weighting(f)
17189 B-weighting of equal loudness
17190 @item c_weighting(f)
17191 C-weighting of equal loudness.
17193 Default value is @code{16}.
17195 @item bar_v, volume2
17196 Specify the bargraph volume expression. It can contain variables:
17199 the @var{sono_v} evaluated expression
17200 @item frequency, freq, f
17201 the frequency where it is evaluated
17202 @item timeclamp, tc
17203 the value of @var{timeclamp} option
17207 @item a_weighting(f)
17208 A-weighting of equal loudness
17209 @item b_weighting(f)
17210 B-weighting of equal loudness
17211 @item c_weighting(f)
17212 C-weighting of equal loudness.
17214 Default value is @code{sono_v}.
17216 @item sono_g, gamma
17217 Specify the sonogram gamma. Lower gamma makes the spectrum more contrast,
17218 higher gamma makes the spectrum having more range. Default value is @code{3}.
17219 Acceptable range is @code{[1, 7]}.
17221 @item bar_g, gamma2
17222 Specify the bargraph gamma. Default value is @code{1}. Acceptable range is
17226 Specify the bargraph transparency level. Lower value makes the bargraph sharper.
17227 Default value is @code{1}. Acceptable range is @code{[0, 1]}.
17229 @item timeclamp, tc
17230 Specify the transform timeclamp. At low frequency, there is trade-off between
17231 accuracy in time domain and frequency domain. If timeclamp is lower,
17232 event in time domain is represented more accurately (such as fast bass drum),
17233 otherwise event in frequency domain is represented more accurately
17234 (such as bass guitar). Acceptable range is @code{[0.002, 1]}. Default value is @code{0.17}.
17237 Set attack time in seconds. The default is @code{0} (disabled). Otherwise, it
17238 limits future samples by applying asymmetric windowing in time domain, useful
17239 when low latency is required. Accepted range is @code{[0, 1]}.
17242 Specify the transform base frequency. Default value is @code{20.01523126408007475},
17243 which is frequency 50 cents below E0. Acceptable range is @code{[10, 100000]}.
17246 Specify the transform end frequency. Default value is @code{20495.59681441799654},
17247 which is frequency 50 cents above D#10. Acceptable range is @code{[10, 100000]}.
17250 This option is deprecated and ignored.
17253 Specify the transform length in time domain. Use this option to control accuracy
17254 trade-off between time domain and frequency domain at every frequency sample.
17255 It can contain variables:
17257 @item frequency, freq, f
17258 the frequency where it is evaluated
17259 @item timeclamp, tc
17260 the value of @var{timeclamp} option.
17262 Default value is @code{384*tc/(384+tc*f)}.
17265 Specify the transform count for every video frame. Default value is @code{6}.
17266 Acceptable range is @code{[1, 30]}.
17269 Specify the transform count for every single pixel. Default value is @code{0},
17270 which makes it computed automatically. Acceptable range is @code{[0, 10]}.
17273 Specify font file for use with freetype to draw the axis. If not specified,
17274 use embedded font. Note that drawing with font file or embedded font is not
17275 implemented with custom @var{basefreq} and @var{endfreq}, use @var{axisfile}
17279 Specify fontconfig pattern. This has lower priority than @var{fontfile}.
17280 The : in the pattern may be replaced by | to avoid unnecessary escaping.
17283 Specify font color expression. This is arithmetic expression that should return
17284 integer value 0xRRGGBB. It can contain variables:
17286 @item frequency, freq, f
17287 the frequency where it is evaluated
17288 @item timeclamp, tc
17289 the value of @var{timeclamp} option
17294 midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
17295 @item r(x), g(x), b(x)
17296 red, green, and blue value of intensity x.
17298 Default value is @code{st(0, (midi(f)-59.5)/12);
17299 st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
17300 r(1-ld(1)) + b(ld(1))}.
17303 Specify image file to draw the axis. This option override @var{fontfile} and
17304 @var{fontcolor} option.
17307 Enable/disable drawing text to the axis. If it is set to @code{0}, drawing to
17308 the axis is disabled, ignoring @var{fontfile} and @var{axisfile} option.
17309 Default value is @code{1}.
17312 Set colorspace. The accepted values are:
17315 Unspecified (default)
17324 BT.470BG or BT.601-6 625
17327 SMPTE-170M or BT.601-6 525
17333 BT.2020 with non-constant luminance
17338 Set spectrogram color scheme. This is list of floating point values with format
17339 @code{left_r|left_g|left_b|right_r|right_g|right_b}.
17340 The default is @code{1|0.5|0|0|0.5|1}.
17344 @subsection Examples
17348 Playing audio while showing the spectrum:
17350 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
17354 Same as above, but with frame rate 30 fps:
17356 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
17360 Playing at 1280x720:
17362 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=s=1280x720:count=4 [out0]'
17366 Disable sonogram display:
17372 A1 and its harmonics: A1, A2, (near)E3, A3:
17374 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),
17375 asplit[a][out1]; [a] showcqt [out0]'
17379 Same as above, but with more accuracy in frequency domain:
17381 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),
17382 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
17388 bar_v=10:sono_v=bar_v*a_weighting(f)
17392 Custom gamma, now spectrum is linear to the amplitude.
17398 Custom tlength equation:
17400 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)))'
17404 Custom fontcolor and fontfile, C-note is colored green, others are colored blue:
17406 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))':fontfile=myfont.ttf
17410 Custom font using fontconfig:
17412 font='Courier New,Monospace,mono|bold'
17416 Custom frequency range with custom axis using image file:
17418 axisfile=myaxis.png:basefreq=40:endfreq=10000
17424 Convert input audio to video output representing the audio power spectrum.
17425 Audio amplitude is on Y-axis while frequency is on X-axis.
17427 The filter accepts the following options:
17431 Specify size of video. For the syntax of this option, check the
17432 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17433 Default is @code{1024x512}.
17437 This set how each frequency bin will be represented.
17439 It accepts the following values:
17445 Default is @code{bar}.
17448 Set amplitude scale.
17450 It accepts the following values:
17464 Default is @code{log}.
17467 Set frequency scale.
17469 It accepts the following values:
17478 Reverse logarithmic scale.
17480 Default is @code{lin}.
17485 It accepts the following values:
17501 Default is @code{w2048}
17504 Set windowing function.
17506 It accepts the following values:
17528 Default is @code{hanning}.
17531 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
17532 which means optimal overlap for selected window function will be picked.
17535 Set time averaging. Setting this to 0 will display current maximal peaks.
17536 Default is @code{1}, which means time averaging is disabled.
17539 Specify list of colors separated by space or by '|' which will be used to
17540 draw channel frequencies. Unrecognized or missing colors will be replaced
17544 Set channel display mode.
17546 It accepts the following values:
17551 Default is @code{combined}.
17554 Set minimum amplitude used in @code{log} amplitude scaler.
17558 @anchor{showspectrum}
17559 @section showspectrum
17561 Convert input audio to a video output, representing the audio frequency
17564 The filter accepts the following options:
17568 Specify the video size for the output. For the syntax of this option, check the
17569 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17570 Default value is @code{640x512}.
17573 Specify how the spectrum should slide along the window.
17575 It accepts the following values:
17578 the samples start again on the left when they reach the right
17580 the samples scroll from right to left
17582 frames are only produced when the samples reach the right
17584 the samples scroll from left to right
17587 Default value is @code{replace}.
17590 Specify display mode.
17592 It accepts the following values:
17595 all channels are displayed in the same row
17597 all channels are displayed in separate rows
17600 Default value is @samp{combined}.
17603 Specify display color mode.
17605 It accepts the following values:
17608 each channel is displayed in a separate color
17610 each channel is displayed using the same color scheme
17612 each channel is displayed using the rainbow color scheme
17614 each channel is displayed using the moreland color scheme
17616 each channel is displayed using the nebulae color scheme
17618 each channel is displayed using the fire color scheme
17620 each channel is displayed using the fiery color scheme
17622 each channel is displayed using the fruit color scheme
17624 each channel is displayed using the cool color scheme
17627 Default value is @samp{channel}.
17630 Specify scale used for calculating intensity color values.
17632 It accepts the following values:
17637 square root, default
17648 Default value is @samp{sqrt}.
17651 Set saturation modifier for displayed colors. Negative values provide
17652 alternative color scheme. @code{0} is no saturation at all.
17653 Saturation must be in [-10.0, 10.0] range.
17654 Default value is @code{1}.
17657 Set window function.
17659 It accepts the following values:
17683 Default value is @code{hann}.
17686 Set orientation of time vs frequency axis. Can be @code{vertical} or
17687 @code{horizontal}. Default is @code{vertical}.
17690 Set ratio of overlap window. Default value is @code{0}.
17691 When value is @code{1} overlap is set to recommended size for specific
17692 window function currently used.
17695 Set scale gain for calculating intensity color values.
17696 Default value is @code{1}.
17699 Set which data to display. Can be @code{magnitude}, default or @code{phase}.
17702 Set color rotation, must be in [-1.0, 1.0] range.
17703 Default value is @code{0}.
17706 The usage is very similar to the showwaves filter; see the examples in that
17709 @subsection Examples
17713 Large window with logarithmic color scaling:
17715 showspectrum=s=1280x480:scale=log
17719 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
17721 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
17722 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
17726 @section showspectrumpic
17728 Convert input audio to a single video frame, representing the audio frequency
17731 The filter accepts the following options:
17735 Specify the video size for the output. For the syntax of this option, check the
17736 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17737 Default value is @code{4096x2048}.
17740 Specify display mode.
17742 It accepts the following values:
17745 all channels are displayed in the same row
17747 all channels are displayed in separate rows
17749 Default value is @samp{combined}.
17752 Specify display color mode.
17754 It accepts the following values:
17757 each channel is displayed in a separate color
17759 each channel is displayed using the same color scheme
17761 each channel is displayed using the rainbow color scheme
17763 each channel is displayed using the moreland color scheme
17765 each channel is displayed using the nebulae color scheme
17767 each channel is displayed using the fire color scheme
17769 each channel is displayed using the fiery color scheme
17771 each channel is displayed using the fruit color scheme
17773 each channel is displayed using the cool color scheme
17775 Default value is @samp{intensity}.
17778 Specify scale used for calculating intensity color values.
17780 It accepts the following values:
17785 square root, default
17795 Default value is @samp{log}.
17798 Set saturation modifier for displayed colors. Negative values provide
17799 alternative color scheme. @code{0} is no saturation at all.
17800 Saturation must be in [-10.0, 10.0] range.
17801 Default value is @code{1}.
17804 Set window function.
17806 It accepts the following values:
17829 Default value is @code{hann}.
17832 Set orientation of time vs frequency axis. Can be @code{vertical} or
17833 @code{horizontal}. Default is @code{vertical}.
17836 Set scale gain for calculating intensity color values.
17837 Default value is @code{1}.
17840 Draw time and frequency axes and legends. Default is enabled.
17843 Set color rotation, must be in [-1.0, 1.0] range.
17844 Default value is @code{0}.
17847 @subsection Examples
17851 Extract an audio spectrogram of a whole audio track
17852 in a 1024x1024 picture using @command{ffmpeg}:
17854 ffmpeg -i audio.flac -lavfi showspectrumpic=s=1024x1024 spectrogram.png
17858 @section showvolume
17860 Convert input audio volume to a video output.
17862 The filter accepts the following options:
17869 Set border width, allowed range is [0, 5]. Default is 1.
17872 Set channel width, allowed range is [80, 8192]. Default is 400.
17875 Set channel height, allowed range is [1, 900]. Default is 20.
17878 Set fade, allowed range is [0.001, 1]. Default is 0.95.
17881 Set volume color expression.
17883 The expression can use the following variables:
17887 Current max volume of channel in dB.
17893 Current channel number, starting from 0.
17897 If set, displays channel names. Default is enabled.
17900 If set, displays volume values. Default is enabled.
17903 Set orientation, can be @code{horizontal} or @code{vertical},
17904 default is @code{horizontal}.
17907 Set step size, allowed range s [0, 5]. Default is 0, which means
17913 Convert input audio to a video output, representing the samples waves.
17915 The filter accepts the following options:
17919 Specify the video size for the output. For the syntax of this option, check the
17920 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
17921 Default value is @code{600x240}.
17926 Available values are:
17929 Draw a point for each sample.
17932 Draw a vertical line for each sample.
17935 Draw a point for each sample and a line between them.
17938 Draw a centered vertical line for each sample.
17941 Default value is @code{point}.
17944 Set the number of samples which are printed on the same column. A
17945 larger value will decrease the frame rate. Must be a positive
17946 integer. This option can be set only if the value for @var{rate}
17947 is not explicitly specified.
17950 Set the (approximate) output frame rate. This is done by setting the
17951 option @var{n}. Default value is "25".
17953 @item split_channels
17954 Set if channels should be drawn separately or overlap. Default value is 0.
17957 Set colors separated by '|' which are going to be used for drawing of each channel.
17960 Set amplitude scale.
17962 Available values are:
17980 @subsection Examples
17984 Output the input file audio and the corresponding video representation
17987 amovie=a.mp3,asplit[out0],showwaves[out1]
17991 Create a synthetic signal and show it with showwaves, forcing a
17992 frame rate of 30 frames per second:
17994 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
17998 @section showwavespic
18000 Convert input audio to a single video frame, representing the samples waves.
18002 The filter accepts the following options:
18006 Specify the video size for the output. For the syntax of this option, check the
18007 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
18008 Default value is @code{600x240}.
18010 @item split_channels
18011 Set if channels should be drawn separately or overlap. Default value is 0.
18014 Set colors separated by '|' which are going to be used for drawing of each channel.
18017 Set amplitude scale.
18019 Available values are:
18037 @subsection Examples
18041 Extract a channel split representation of the wave form of a whole audio track
18042 in a 1024x800 picture using @command{ffmpeg}:
18044 ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
18048 @section sidedata, asidedata
18050 Delete frame side data, or select frames based on it.
18052 This filter accepts the following options:
18056 Set mode of operation of the filter.
18058 Can be one of the following:
18062 Select every frame with side data of @code{type}.
18065 Delete side data of @code{type}. If @code{type} is not set, delete all side
18071 Set side data type used with all modes. Must be set for @code{select} mode. For
18072 the list of frame side data types, refer to the @code{AVFrameSideDataType} enum
18073 in @file{libavutil/frame.h}. For example, to choose
18074 @code{AV_FRAME_DATA_PANSCAN} side data, you must specify @code{PANSCAN}.
18078 @section spectrumsynth
18080 Sythesize audio from 2 input video spectrums, first input stream represents
18081 magnitude across time and second represents phase across time.
18082 The filter will transform from frequency domain as displayed in videos back
18083 to time domain as presented in audio output.
18085 This filter is primarily created for reversing processed @ref{showspectrum}
18086 filter outputs, but can synthesize sound from other spectrograms too.
18087 But in such case results are going to be poor if the phase data is not
18088 available, because in such cases phase data need to be recreated, usually
18089 its just recreated from random noise.
18090 For best results use gray only output (@code{channel} color mode in
18091 @ref{showspectrum} filter) and @code{log} scale for magnitude video and
18092 @code{lin} scale for phase video. To produce phase, for 2nd video, use
18093 @code{data} option. Inputs videos should generally use @code{fullframe}
18094 slide mode as that saves resources needed for decoding video.
18096 The filter accepts the following options:
18100 Specify sample rate of output audio, the sample rate of audio from which
18101 spectrum was generated may differ.
18104 Set number of channels represented in input video spectrums.
18107 Set scale which was used when generating magnitude input spectrum.
18108 Can be @code{lin} or @code{log}. Default is @code{log}.
18111 Set slide which was used when generating inputs spectrums.
18112 Can be @code{replace}, @code{scroll}, @code{fullframe} or @code{rscroll}.
18113 Default is @code{fullframe}.
18116 Set window function used for resynthesis.
18119 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
18120 which means optimal overlap for selected window function will be picked.
18123 Set orientation of input videos. Can be @code{vertical} or @code{horizontal}.
18124 Default is @code{vertical}.
18127 @subsection Examples
18131 First create magnitude and phase videos from audio, assuming audio is stereo with 44100 sample rate,
18132 then resynthesize videos back to audio with spectrumsynth:
18134 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
18135 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
18136 ffmpeg -i magnitude.nut -i phase.nut -lavfi spectrumsynth=channels=2:sample_rate=44100:win_func=hann:overlap=0.875:slide=fullframe output.flac
18140 @section split, asplit
18142 Split input into several identical outputs.
18144 @code{asplit} works with audio input, @code{split} with video.
18146 The filter accepts a single parameter which specifies the number of outputs. If
18147 unspecified, it defaults to 2.
18149 @subsection Examples
18153 Create two separate outputs from the same input:
18155 [in] split [out0][out1]
18159 To create 3 or more outputs, you need to specify the number of
18162 [in] asplit=3 [out0][out1][out2]
18166 Create two separate outputs from the same input, one cropped and
18169 [in] split [splitout1][splitout2];
18170 [splitout1] crop=100:100:0:0 [cropout];
18171 [splitout2] pad=200:200:100:100 [padout];
18175 Create 5 copies of the input audio with @command{ffmpeg}:
18177 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
18183 Receive commands sent through a libzmq client, and forward them to
18184 filters in the filtergraph.
18186 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
18187 must be inserted between two video filters, @code{azmq} between two
18190 To enable these filters you need to install the libzmq library and
18191 headers and configure FFmpeg with @code{--enable-libzmq}.
18193 For more information about libzmq see:
18194 @url{http://www.zeromq.org/}
18196 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
18197 receives messages sent through a network interface defined by the
18198 @option{bind_address} option.
18200 The received message must be in the form:
18202 @var{TARGET} @var{COMMAND} [@var{ARG}]
18205 @var{TARGET} specifies the target of the command, usually the name of
18206 the filter class or a specific filter instance name.
18208 @var{COMMAND} specifies the name of the command for the target filter.
18210 @var{ARG} is optional and specifies the optional argument list for the
18211 given @var{COMMAND}.
18213 Upon reception, the message is processed and the corresponding command
18214 is injected into the filtergraph. Depending on the result, the filter
18215 will send a reply to the client, adopting the format:
18217 @var{ERROR_CODE} @var{ERROR_REASON}
18221 @var{MESSAGE} is optional.
18223 @subsection Examples
18225 Look at @file{tools/zmqsend} for an example of a zmq client which can
18226 be used to send commands processed by these filters.
18228 Consider the following filtergraph generated by @command{ffplay}
18230 ffplay -dumpgraph 1 -f lavfi "
18231 color=s=100x100:c=red [l];
18232 color=s=100x100:c=blue [r];
18233 nullsrc=s=200x100, zmq [bg];
18234 [bg][l] overlay [bg+l];
18235 [bg+l][r] overlay=x=100 "
18238 To change the color of the left side of the video, the following
18239 command can be used:
18241 echo Parsed_color_0 c yellow | tools/zmqsend
18244 To change the right side:
18246 echo Parsed_color_1 c pink | tools/zmqsend
18249 @c man end MULTIMEDIA FILTERS
18251 @chapter Multimedia Sources
18252 @c man begin MULTIMEDIA SOURCES
18254 Below is a description of the currently available multimedia sources.
18258 This is the same as @ref{movie} source, except it selects an audio
18264 Read audio and/or video stream(s) from a movie container.
18266 It accepts the following parameters:
18270 The name of the resource to read (not necessarily a file; it can also be a
18271 device or a stream accessed through some protocol).
18273 @item format_name, f
18274 Specifies the format assumed for the movie to read, and can be either
18275 the name of a container or an input device. If not specified, the
18276 format is guessed from @var{movie_name} or by probing.
18278 @item seek_point, sp
18279 Specifies the seek point in seconds. The frames will be output
18280 starting from this seek point. The parameter is evaluated with
18281 @code{av_strtod}, so the numerical value may be suffixed by an IS
18282 postfix. The default value is "0".
18285 Specifies the streams to read. Several streams can be specified,
18286 separated by "+". The source will then have as many outputs, in the
18287 same order. The syntax is explained in the ``Stream specifiers''
18288 section in the ffmpeg manual. Two special names, "dv" and "da" specify
18289 respectively the default (best suited) video and audio stream. Default
18290 is "dv", or "da" if the filter is called as "amovie".
18292 @item stream_index, si
18293 Specifies the index of the video stream to read. If the value is -1,
18294 the most suitable video stream will be automatically selected. The default
18295 value is "-1". Deprecated. If the filter is called "amovie", it will select
18296 audio instead of video.
18299 Specifies how many times to read the stream in sequence.
18300 If the value is 0, the stream will be looped infinitely.
18301 Default value is "1".
18303 Note that when the movie is looped the source timestamps are not
18304 changed, so it will generate non monotonically increasing timestamps.
18306 @item discontinuity
18307 Specifies the time difference between frames above which the point is
18308 considered a timestamp discontinuity which is removed by adjusting the later
18312 It allows overlaying a second video on top of the main input of
18313 a filtergraph, as shown in this graph:
18315 input -----------> deltapts0 --> overlay --> output
18318 movie --> scale--> deltapts1 -------+
18320 @subsection Examples
18324 Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
18325 on top of the input labelled "in":
18327 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
18328 [in] setpts=PTS-STARTPTS [main];
18329 [main][over] overlay=16:16 [out]
18333 Read from a video4linux2 device, and overlay it on top of the input
18336 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
18337 [in] setpts=PTS-STARTPTS [main];
18338 [main][over] overlay=16:16 [out]
18342 Read the first video stream and the audio stream with id 0x81 from
18343 dvd.vob; the video is connected to the pad named "video" and the audio is
18344 connected to the pad named "audio":
18346 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
18350 @subsection Commands
18352 Both movie and amovie support the following commands:
18355 Perform seek using "av_seek_frame".
18356 The syntax is: seek @var{stream_index}|@var{timestamp}|@var{flags}
18359 @var{stream_index}: If stream_index is -1, a default
18360 stream is selected, and @var{timestamp} is automatically converted
18361 from AV_TIME_BASE units to the stream specific time_base.
18363 @var{timestamp}: Timestamp in AVStream.time_base units
18364 or, if no stream is specified, in AV_TIME_BASE units.
18366 @var{flags}: Flags which select direction and seeking mode.
18370 Get movie duration in AV_TIME_BASE units.
18374 @c man end MULTIMEDIA SOURCES