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 @c man end FILTERGRAPH DESCRIPTION
311 @chapter Audio Filters
312 @c man begin AUDIO FILTERS
314 When you configure your FFmpeg build, you can disable any of the
315 existing filters using @code{--disable-filters}.
316 The configure output will show the audio filters included in your
319 Below is a description of the currently available audio filters.
323 A compressor is mainly used to reduce the dynamic range of a signal.
324 Especially modern music is mostly compressed at a high ratio to
325 improve the overall loudness. It's done to get the highest attention
326 of a listener, "fatten" the sound and bring more "power" to the track.
327 If a signal is compressed too much it may sound dull or "dead"
328 afterwards or it may start to "pump" (which could be a powerful effect
329 but can also destroy a track completely).
330 The right compression is the key to reach a professional sound and is
331 the high art of mixing and mastering. Because of its complex settings
332 it may take a long time to get the right feeling for this kind of effect.
334 Compression is done by detecting the volume above a chosen level
335 @code{threshold} and dividing it by the factor set with @code{ratio}.
336 So if you set the threshold to -12dB and your signal reaches -6dB a ratio
337 of 2:1 will result in a signal at -9dB. Because an exact manipulation of
338 the signal would cause distortion of the waveform the reduction can be
339 levelled over the time. This is done by setting "Attack" and "Release".
340 @code{attack} determines how long the signal has to rise above the threshold
341 before any reduction will occur and @code{release} sets the time the signal
342 has to fall below the threshold to reduce the reduction again. Shorter signals
343 than the chosen attack time will be left untouched.
344 The overall reduction of the signal can be made up afterwards with the
345 @code{makeup} setting. So compressing the peaks of a signal about 6dB and
346 raising the makeup to this level results in a signal twice as loud than the
347 source. To gain a softer entry in the compression the @code{knee} flattens the
348 hard edge at the threshold in the range of the chosen decibels.
350 The filter accepts the following options:
354 Set input gain. Default is 1. Range is between 0.015625 and 64.
357 If a signal of second stream rises above this level it will affect the gain
358 reduction of the first stream.
359 By default it is 0.125. Range is between 0.00097563 and 1.
362 Set a ratio by which the signal is reduced. 1:2 means that if the level
363 rose 4dB above the threshold, it will be only 2dB above after the reduction.
364 Default is 2. Range is between 1 and 20.
367 Amount of milliseconds the signal has to rise above the threshold before gain
368 reduction starts. Default is 20. Range is between 0.01 and 2000.
371 Amount of milliseconds the signal has to fall below the threshold before
372 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
375 Set the amount by how much signal will be amplified after processing.
376 Default is 2. Range is from 1 and 64.
379 Curve the sharp knee around the threshold to enter gain reduction more softly.
380 Default is 2.82843. Range is between 1 and 8.
383 Choose if the @code{average} level between all channels of input stream
384 or the louder(@code{maximum}) channel of input stream affects the
385 reduction. Default is @code{average}.
388 Should the exact signal be taken in case of @code{peak} or an RMS one in case
389 of @code{rms}. Default is @code{rms} which is mostly smoother.
392 How much to use compressed signal in output. Default is 1.
393 Range is between 0 and 1.
398 Apply cross fade from one input audio stream to another input audio stream.
399 The cross fade is applied for specified duration near the end of first stream.
401 The filter accepts the following options:
405 Specify the number of samples for which the cross fade effect has to last.
406 At the end of the cross fade effect the first input audio will be completely
407 silent. Default is 44100.
410 Specify the duration of the cross fade effect. See
411 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
412 for the accepted syntax.
413 By default the duration is determined by @var{nb_samples}.
414 If set this option is used instead of @var{nb_samples}.
417 Should first stream end overlap with second stream start. Default is enabled.
420 Set curve for cross fade transition for first stream.
423 Set curve for cross fade transition for second stream.
425 For description of available curve types see @ref{afade} filter description.
432 Cross fade from one input to another:
434 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:c1=exp:c2=exp output.flac
438 Cross fade from one input to another but without overlapping:
440 ffmpeg -i first.flac -i second.flac -filter_complex acrossfade=d=10:o=0:c1=exp:c2=exp output.flac
446 Delay one or more audio channels.
448 Samples in delayed channel are filled with silence.
450 The filter accepts the following option:
454 Set list of delays in milliseconds for each channel separated by '|'.
455 At least one delay greater than 0 should be provided.
456 Unused delays will be silently ignored. If number of given delays is
457 smaller than number of channels all remaining channels will not be delayed.
464 Delay first channel by 1.5 seconds, the third channel by 0.5 seconds and leave
465 the second channel (and any other channels that may be present) unchanged.
473 Apply echoing to the input audio.
475 Echoes are reflected sound and can occur naturally amongst mountains
476 (and sometimes large buildings) when talking or shouting; digital echo
477 effects emulate this behaviour and are often used to help fill out the
478 sound of a single instrument or vocal. The time difference between the
479 original signal and the reflection is the @code{delay}, and the
480 loudness of the reflected signal is the @code{decay}.
481 Multiple echoes can have different delays and decays.
483 A description of the accepted parameters follows.
487 Set input gain of reflected signal. Default is @code{0.6}.
490 Set output gain of reflected signal. Default is @code{0.3}.
493 Set list of time intervals in milliseconds between original signal and reflections
494 separated by '|'. Allowed range for each @code{delay} is @code{(0 - 90000.0]}.
495 Default is @code{1000}.
498 Set list of loudnesses of reflected signals separated by '|'.
499 Allowed range for each @code{decay} is @code{(0 - 1.0]}.
500 Default is @code{0.5}.
507 Make it sound as if there are twice as many instruments as are actually playing:
509 aecho=0.8:0.88:60:0.4
513 If delay is very short, then it sound like a (metallic) robot playing music:
519 A longer delay will sound like an open air concert in the mountains:
521 aecho=0.8:0.9:1000:0.3
525 Same as above but with one more mountain:
527 aecho=0.8:0.9:1000|1800:0.3|0.25
532 Audio emphasis filter creates or restores material directly taken from LPs or
533 emphased CDs with different filter curves. E.g. to store music on vinyl the
534 signal has to be altered by a filter first to even out the disadvantages of
535 this recording medium.
536 Once the material is played back the inverse filter has to be applied to
537 restore the distortion of the frequency response.
539 The filter accepts the following options:
549 Set filter mode. For restoring material use @code{reproduction} mode, otherwise
550 use @code{production} mode. Default is @code{reproduction} mode.
553 Set filter type. Selects medium. Can be one of the following:
565 select Compact Disc (CD).
571 select 50µs (FM-KF).
573 select 75µs (FM-KF).
579 Modify an audio signal according to the specified expressions.
581 This filter accepts one or more expressions (one for each channel),
582 which are evaluated and used to modify a corresponding audio signal.
584 It accepts the following parameters:
588 Set the '|'-separated expressions list for each separate channel. If
589 the number of input channels is greater than the number of
590 expressions, the last specified expression is used for the remaining
593 @item channel_layout, c
594 Set output channel layout. If not specified, the channel layout is
595 specified by the number of expressions. If set to @samp{same}, it will
596 use by default the same input channel layout.
599 Each expression in @var{exprs} can contain the following constants and functions:
603 channel number of the current expression
606 number of the evaluated sample, starting from 0
612 time of the evaluated sample expressed in seconds
615 @item nb_out_channels
616 input and output number of channels
619 the value of input channel with number @var{CH}
622 Note: this filter is slow. For faster processing you should use a
631 aeval=val(ch)/2:c=same
635 Invert phase of the second channel:
644 Apply fade-in/out effect to input audio.
646 A description of the accepted parameters follows.
650 Specify the effect type, can be either @code{in} for fade-in, or
651 @code{out} for a fade-out effect. Default is @code{in}.
653 @item start_sample, ss
654 Specify the number of the start sample for starting to apply the fade
655 effect. Default is 0.
658 Specify the number of samples for which the fade effect has to last. At
659 the end of the fade-in effect the output audio will have the same
660 volume as the input audio, at the end of the fade-out transition
661 the output audio will be silence. Default is 44100.
664 Specify the start time of the fade effect. Default is 0.
665 The value must be specified as a time duration; see
666 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
667 for the accepted syntax.
668 If set this option is used instead of @var{start_sample}.
671 Specify the duration of the fade effect. See
672 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
673 for the accepted syntax.
674 At the end of the fade-in effect the output audio will have the same
675 volume as the input audio, at the end of the fade-out transition
676 the output audio will be silence.
677 By default the duration is determined by @var{nb_samples}.
678 If set this option is used instead of @var{nb_samples}.
681 Set curve for fade transition.
683 It accepts the following values:
686 select triangular, linear slope (default)
688 select quarter of sine wave
690 select half of sine wave
692 select exponential sine wave
696 select inverted parabola
710 select inverted quarter of sine wave
712 select inverted half of sine wave
714 select double-exponential seat
716 select double-exponential sigmoid
724 Fade in first 15 seconds of audio:
730 Fade out last 25 seconds of a 900 seconds audio:
732 afade=t=out:st=875:d=25
737 Apply arbitrary expressions to samples in frequency domain.
741 Set frequency domain real expression for each separate channel separated
742 by '|'. Default is "1".
743 If the number of input channels is greater than the number of
744 expressions, the last specified expression is used for the remaining
748 Set frequency domain imaginary expression for each separate channel
749 separated by '|'. If not set, @var{real} option is used.
751 Each expression in @var{real} and @var{imag} can contain the following
759 current frequency bin number
762 number of available bins
765 channel number of the current expression
777 It accepts the following values:
793 Default is @code{w4096}
796 Set window function. Default is @code{hann}.
799 Set window overlap. If set to 1, the recommended overlap for selected
800 window function will be picked. Default is @code{0.75}.
807 Leave almost only low frequencies in audio:
809 afftfilt="1-clip((b/nb)*b,0,1)"
816 Set output format constraints for the input audio. The framework will
817 negotiate the most appropriate format to minimize conversions.
819 It accepts the following parameters:
823 A '|'-separated list of requested sample formats.
826 A '|'-separated list of requested sample rates.
828 @item channel_layouts
829 A '|'-separated list of requested channel layouts.
831 See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
832 for the required syntax.
835 If a parameter is omitted, all values are allowed.
837 Force the output to either unsigned 8-bit or signed 16-bit stereo
839 aformat=sample_fmts=u8|s16:channel_layouts=stereo
844 A gate is mainly used to reduce lower parts of a signal. This kind of signal
845 processing reduces disturbing noise between useful signals.
847 Gating is done by detecting the volume below a chosen level @var{threshold}
848 and divide it by the factor set with @var{ratio}. The bottom of the noise
849 floor is set via @var{range}. Because an exact manipulation of the signal
850 would cause distortion of the waveform the reduction can be levelled over
851 time. This is done by setting @var{attack} and @var{release}.
853 @var{attack} determines how long the signal has to fall below the threshold
854 before any reduction will occur and @var{release} sets the time the signal
855 has to raise above the threshold to reduce the reduction again.
856 Shorter signals than the chosen attack time will be left untouched.
860 Set input level before filtering.
861 Default is 1. Allowed range is from 0.015625 to 64.
864 Set the level of gain reduction when the signal is below the threshold.
865 Default is 0.06125. Allowed range is from 0 to 1.
868 If a signal rises above this level the gain reduction is released.
869 Default is 0.125. Allowed range is from 0 to 1.
872 Set a ratio about which the signal is reduced.
873 Default is 2. Allowed range is from 1 to 9000.
876 Amount of milliseconds the signal has to rise above the threshold before gain
878 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
881 Amount of milliseconds the signal has to fall below the threshold before the
882 reduction is increased again. Default is 250 milliseconds.
883 Allowed range is from 0.01 to 9000.
886 Set amount of amplification of signal after processing.
887 Default is 1. Allowed range is from 1 to 64.
890 Curve the sharp knee around the threshold to enter gain reduction more softly.
891 Default is 2.828427125. Allowed range is from 1 to 8.
894 Choose if exact signal should be taken for detection or an RMS like one.
895 Default is rms. Can be peak or rms.
898 Choose if the average level between all channels or the louder channel affects
900 Default is average. Can be average or maximum.
905 The limiter prevents input signal from raising over a desired threshold.
906 This limiter uses lookahead technology to prevent your signal from distorting.
907 It means that there is a small delay after signal is processed. Keep in mind
908 that the delay it produces is the attack time you set.
910 The filter accepts the following options:
914 Set input gain. Default is 1.
917 Set output gain. Default is 1.
920 Don't let signals above this level pass the limiter. Default is 1.
923 The limiter will reach its attenuation level in this amount of time in
924 milliseconds. Default is 5 milliseconds.
927 Come back from limiting to attenuation 1.0 in this amount of milliseconds.
928 Default is 50 milliseconds.
931 When gain reduction is always needed ASC takes care of releasing to an
932 average reduction level rather than reaching a reduction of 0 in the release
936 Select how much the release time is affected by ASC, 0 means nearly no changes
937 in release time while 1 produces higher release times.
940 Auto level output signal. Default is enabled.
941 This normalizes audio back to 0dB if enabled.
944 Depending on picked setting it is recommended to upsample input 2x or 4x times
945 with @ref{aresample} before applying this filter.
949 Apply a two-pole all-pass filter with central frequency (in Hz)
950 @var{frequency}, and filter-width @var{width}.
951 An all-pass filter changes the audio's frequency to phase relationship
952 without changing its frequency to amplitude relationship.
954 The filter accepts the following options:
961 Set method to specify band-width of filter.
974 Specify the band-width of a filter in width_type units.
980 Merge two or more audio streams into a single multi-channel stream.
982 The filter accepts the following options:
987 Set the number of inputs. Default is 2.
991 If the channel layouts of the inputs are disjoint, and therefore compatible,
992 the channel layout of the output will be set accordingly and the channels
993 will be reordered as necessary. If the channel layouts of the inputs are not
994 disjoint, the output will have all the channels of the first input then all
995 the channels of the second input, in that order, and the channel layout of
996 the output will be the default value corresponding to the total number of
999 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
1000 is FC+BL+BR, then the output will be in 5.1, with the channels in the
1001 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
1002 first input, b1 is the first channel of the second input).
1004 On the other hand, if both input are in stereo, the output channels will be
1005 in the default order: a1, a2, b1, b2, and the channel layout will be
1006 arbitrarily set to 4.0, which may or may not be the expected value.
1008 All inputs must have the same sample rate, and format.
1010 If inputs do not have the same duration, the output will stop with the
1013 @subsection Examples
1017 Merge two mono files into a stereo stream:
1019 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
1023 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
1025 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
1031 Mixes multiple audio inputs into a single output.
1033 Note that this filter only supports float samples (the @var{amerge}
1034 and @var{pan} audio filters support many formats). If the @var{amix}
1035 input has integer samples then @ref{aresample} will be automatically
1036 inserted to perform the conversion to float samples.
1040 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
1042 will mix 3 input audio streams to a single output with the same duration as the
1043 first input and a dropout transition time of 3 seconds.
1045 It accepts the following parameters:
1049 The number of inputs. If unspecified, it defaults to 2.
1052 How to determine the end-of-stream.
1056 The duration of the longest input. (default)
1059 The duration of the shortest input.
1062 The duration of the first input.
1066 @item dropout_transition
1067 The transition time, in seconds, for volume renormalization when an input
1068 stream ends. The default value is 2 seconds.
1072 @section anequalizer
1074 High-order parametric multiband equalizer for each channel.
1076 It accepts the following parameters:
1080 This option string is in format:
1081 "c@var{chn} f=@var{cf} w=@var{w} g=@var{g} t=@var{f} | ..."
1082 Each equalizer band is separated by '|'.
1086 Set channel number to which equalization will be applied.
1087 If input doesn't have that channel the entry is ignored.
1090 Set central frequency for band.
1091 If input doesn't have that frequency the entry is ignored.
1094 Set band width in hertz.
1097 Set band gain in dB.
1100 Set filter type for band, optional, can be:
1104 Butterworth, this is default.
1115 With this option activated frequency response of anequalizer is displayed
1119 Set video stream size. Only useful if curves option is activated.
1122 Set max gain that will be displayed. Only useful if curves option is activated.
1123 Setting this to reasonable value allows to display gain which is derived from
1124 neighbour bands which are too close to each other and thus produce higher gain
1125 when both are activated.
1128 Set frequency scale used to draw frequency response in video output.
1129 Can be linear or logarithmic. Default is logarithmic.
1132 Set color for each channel curve which is going to be displayed in video stream.
1133 This is list of color names separated by space or by '|'.
1134 Unrecognised or missing colors will be replaced by white color.
1137 @subsection Examples
1141 Lower gain by 10 of central frequency 200Hz and width 100 Hz
1142 for first 2 channels using Chebyshev type 1 filter:
1144 anequalizer=c0 f=200 w=100 g=-10 t=1|c1 f=200 w=100 g=-10 t=1
1148 @subsection Commands
1150 This filter supports the following commands:
1153 Alter existing filter parameters.
1154 Syntax for the commands is : "@var{fN}|f=@var{freq}|w=@var{width}|g=@var{gain}"
1156 @var{fN} is existing filter number, starting from 0, if no such filter is available
1158 @var{freq} set new frequency parameter.
1159 @var{width} set new width parameter in herz.
1160 @var{gain} set new gain parameter in dB.
1162 Full filter invocation with asendcmd may look like this:
1163 asendcmd=c='4.0 anequalizer change 0|f=200|w=50|g=1',anequalizer=...
1168 Pass the audio source unchanged to the output.
1172 Pad the end of an audio stream with silence.
1174 This can be used together with @command{ffmpeg} @option{-shortest} to
1175 extend audio streams to the same length as the video stream.
1177 A description of the accepted options follows.
1181 Set silence packet size. Default value is 4096.
1184 Set the number of samples of silence to add to the end. After the
1185 value is reached, the stream is terminated. This option is mutually
1186 exclusive with @option{whole_len}.
1189 Set the minimum total number of samples in the output audio stream. If
1190 the value is longer than the input audio length, silence is added to
1191 the end, until the value is reached. This option is mutually exclusive
1192 with @option{pad_len}.
1195 If neither the @option{pad_len} nor the @option{whole_len} option is
1196 set, the filter will add silence to the end of the input stream
1199 @subsection Examples
1203 Add 1024 samples of silence to the end of the input:
1209 Make sure the audio output will contain at least 10000 samples, pad
1210 the input with silence if required:
1212 apad=whole_len=10000
1216 Use @command{ffmpeg} to pad the audio input with silence, so that the
1217 video stream will always result the shortest and will be converted
1218 until the end in the output file when using the @option{shortest}
1221 ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
1226 Add a phasing effect to the input audio.
1228 A phaser filter creates series of peaks and troughs in the frequency spectrum.
1229 The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
1231 A description of the accepted parameters follows.
1235 Set input gain. Default is 0.4.
1238 Set output gain. Default is 0.74
1241 Set delay in milliseconds. Default is 3.0.
1244 Set decay. Default is 0.4.
1247 Set modulation speed in Hz. Default is 0.5.
1250 Set modulation type. Default is triangular.
1252 It accepts the following values:
1261 Audio pulsator is something between an autopanner and a tremolo.
1262 But it can produce funny stereo effects as well. Pulsator changes the volume
1263 of the left and right channel based on a LFO (low frequency oscillator) with
1264 different waveforms and shifted phases.
1265 This filter have the ability to define an offset between left and right
1266 channel. An offset of 0 means that both LFO shapes match each other.
1267 The left and right channel are altered equally - a conventional tremolo.
1268 An offset of 50% means that the shape of the right channel is exactly shifted
1269 in phase (or moved backwards about half of the frequency) - pulsator acts as
1270 an autopanner. At 1 both curves match again. Every setting in between moves the
1271 phase shift gapless between all stages and produces some "bypassing" sounds with
1272 sine and triangle waveforms. The more you set the offset near 1 (starting from
1273 the 0.5) the faster the signal passes from the left to the right speaker.
1275 The filter accepts the following options:
1279 Set input gain. By default it is 1. Range is [0.015625 - 64].
1282 Set output gain. By default it is 1. Range is [0.015625 - 64].
1285 Set waveform shape the LFO will use. Can be one of: sine, triangle, square,
1286 sawup or sawdown. Default is sine.
1289 Set modulation. Define how much of original signal is affected by the LFO.
1292 Set left channel offset. Default is 0. Allowed range is [0 - 1].
1295 Set right channel offset. Default is 0.5. Allowed range is [0 - 1].
1298 Set pulse width. Default is 1. Allowed range is [0 - 2].
1301 Set possible timing mode. Can be one of: bpm, ms or hz. Default is hz.
1304 Set bpm. Default is 120. Allowed range is [30 - 300]. Only used if timing
1308 Set ms. Default is 500. Allowed range is [10 - 2000]. Only used if timing
1312 Set frequency in Hz. Default is 2. Allowed range is [0.01 - 100]. Only used
1313 if timing is set to hz.
1319 Resample the input audio to the specified parameters, using the
1320 libswresample library. If none are specified then the filter will
1321 automatically convert between its input and output.
1323 This filter is also able to stretch/squeeze the audio data to make it match
1324 the timestamps or to inject silence / cut out audio to make it match the
1325 timestamps, do a combination of both or do neither.
1327 The filter accepts the syntax
1328 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
1329 expresses a sample rate and @var{resampler_options} is a list of
1330 @var{key}=@var{value} pairs, separated by ":". See the
1331 ffmpeg-resampler manual for the complete list of supported options.
1333 @subsection Examples
1337 Resample the input audio to 44100Hz:
1343 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
1344 samples per second compensation:
1346 aresample=async=1000
1350 @section asetnsamples
1352 Set the number of samples per each output audio frame.
1354 The last output packet may contain a different number of samples, as
1355 the filter will flush all the remaining samples when the input audio
1358 The filter accepts the following options:
1362 @item nb_out_samples, n
1363 Set the number of frames per each output audio frame. The number is
1364 intended as the number of samples @emph{per each channel}.
1365 Default value is 1024.
1368 If set to 1, the filter will pad the last audio frame with zeroes, so
1369 that the last frame will contain the same number of samples as the
1370 previous ones. Default value is 1.
1373 For example, to set the number of per-frame samples to 1234 and
1374 disable padding for the last frame, use:
1376 asetnsamples=n=1234:p=0
1381 Set the sample rate without altering the PCM data.
1382 This will result in a change of speed and pitch.
1384 The filter accepts the following options:
1387 @item sample_rate, r
1388 Set the output sample rate. Default is 44100 Hz.
1393 Show a line containing various information for each input audio frame.
1394 The input audio is not modified.
1396 The shown line contains a sequence of key/value pairs of the form
1397 @var{key}:@var{value}.
1399 The following values are shown in the output:
1403 The (sequential) number of the input frame, starting from 0.
1406 The presentation timestamp of the input frame, in time base units; the time base
1407 depends on the filter input pad, and is usually 1/@var{sample_rate}.
1410 The presentation timestamp of the input frame in seconds.
1413 position of the frame in the input stream, -1 if this information in
1414 unavailable and/or meaningless (for example in case of synthetic audio)
1423 The sample rate for the audio frame.
1426 The number of samples (per channel) in the frame.
1429 The Adler-32 checksum (printed in hexadecimal) of the audio data. For planar
1430 audio, the data is treated as if all the planes were concatenated.
1432 @item plane_checksums
1433 A list of Adler-32 checksums for each data plane.
1439 Display time domain statistical information about the audio channels.
1440 Statistics are calculated and displayed for each audio channel and,
1441 where applicable, an overall figure is also given.
1443 It accepts the following option:
1446 Short window length in seconds, used for peak and trough RMS measurement.
1447 Default is @code{0.05} (50 milliseconds). Allowed range is @code{[0.1 - 10]}.
1451 Set metadata injection. All the metadata keys are prefixed with @code{lavfi.astats.X},
1452 where @code{X} is channel number starting from 1 or string @code{Overall}. Default is
1455 Available keys for each channel are:
1486 For example full key look like this @code{lavfi.astats.1.DC_offset} or
1487 this @code{lavfi.astats.Overall.Peak_count}.
1489 For description what each key means read below.
1492 Set number of frame after which stats are going to be recalculated.
1493 Default is disabled.
1496 A description of each shown parameter follows:
1500 Mean amplitude displacement from zero.
1503 Minimal sample level.
1506 Maximal sample level.
1508 @item Min difference
1509 Minimal difference between two consecutive samples.
1511 @item Max difference
1512 Maximal difference between two consecutive samples.
1514 @item Mean difference
1515 Mean difference between two consecutive samples.
1516 The average of each difference between two consecutive samples.
1520 Standard peak and RMS level measured in dBFS.
1524 Peak and trough values for RMS level measured over a short window.
1527 Standard ratio of peak to RMS level (note: not in dB).
1530 Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
1531 (i.e. either @var{Min level} or @var{Max level}).
1534 Number of occasions (not the number of samples) that the signal attained either
1535 @var{Min level} or @var{Max level}.
1538 Overall bit depth of audio. Number of bits used for each sample.
1543 Synchronize audio data with timestamps by squeezing/stretching it and/or
1544 dropping samples/adding silence when needed.
1546 This filter is not built by default, please use @ref{aresample} to do squeezing/stretching.
1548 It accepts the following parameters:
1552 Enable stretching/squeezing the data to make it match the timestamps. Disabled
1553 by default. When disabled, time gaps are covered with silence.
1556 The minimum difference between timestamps and audio data (in seconds) to trigger
1557 adding/dropping samples. The default value is 0.1. If you get an imperfect
1558 sync with this filter, try setting this parameter to 0.
1561 The maximum compensation in samples per second. Only relevant with compensate=1.
1562 The default value is 500.
1565 Assume that the first PTS should be this value. The time base is 1 / sample
1566 rate. This allows for padding/trimming at the start of the stream. By default,
1567 no assumption is made about the first frame's expected PTS, so no padding or
1568 trimming is done. For example, this could be set to 0 to pad the beginning with
1569 silence if an audio stream starts after the video stream or to trim any samples
1570 with a negative PTS due to encoder delay.
1578 The filter accepts exactly one parameter, the audio tempo. If not
1579 specified then the filter will assume nominal 1.0 tempo. Tempo must
1580 be in the [0.5, 2.0] range.
1582 @subsection Examples
1586 Slow down audio to 80% tempo:
1592 To speed up audio to 125% tempo:
1600 Trim the input so that the output contains one continuous subpart of the input.
1602 It accepts the following parameters:
1605 Timestamp (in seconds) of the start of the section to keep. I.e. the audio
1606 sample with the timestamp @var{start} will be the first sample in the output.
1609 Specify time of the first audio sample that will be dropped, i.e. the
1610 audio sample immediately preceding the one with the timestamp @var{end} will be
1611 the last sample in the output.
1614 Same as @var{start}, except this option sets the start timestamp in samples
1618 Same as @var{end}, except this option sets the end timestamp in samples instead
1622 The maximum duration of the output in seconds.
1625 The number of the first sample that should be output.
1628 The number of the first sample that should be dropped.
1631 @option{start}, @option{end}, and @option{duration} are expressed as time
1632 duration specifications; see
1633 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
1635 Note that the first two sets of the start/end options and the @option{duration}
1636 option look at the frame timestamp, while the _sample options simply count the
1637 samples that pass through the filter. So start/end_pts and start/end_sample will
1638 give different results when the timestamps are wrong, inexact or do not start at
1639 zero. Also note that this filter does not modify the timestamps. If you wish
1640 to have the output timestamps start at zero, insert the asetpts filter after the
1643 If multiple start or end options are set, this filter tries to be greedy and
1644 keep all samples that match at least one of the specified constraints. To keep
1645 only the part that matches all the constraints at once, chain multiple atrim
1648 The defaults are such that all the input is kept. So it is possible to set e.g.
1649 just the end values to keep everything before the specified time.
1654 Drop everything except the second minute of input:
1656 ffmpeg -i INPUT -af atrim=60:120
1660 Keep only the first 1000 samples:
1662 ffmpeg -i INPUT -af atrim=end_sample=1000
1669 Apply a two-pole Butterworth band-pass filter with central
1670 frequency @var{frequency}, and (3dB-point) band-width width.
1671 The @var{csg} option selects a constant skirt gain (peak gain = Q)
1672 instead of the default: constant 0dB peak gain.
1673 The filter roll off at 6dB per octave (20dB per decade).
1675 The filter accepts the following options:
1679 Set the filter's central frequency. Default is @code{3000}.
1682 Constant skirt gain if set to 1. Defaults to 0.
1685 Set method to specify band-width of filter.
1698 Specify the band-width of a filter in width_type units.
1703 Apply a two-pole Butterworth band-reject filter with central
1704 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1705 The filter roll off at 6dB per octave (20dB per decade).
1707 The filter accepts the following options:
1711 Set the filter's central frequency. Default is @code{3000}.
1714 Set method to specify band-width of filter.
1727 Specify the band-width of a filter in width_type units.
1732 Boost or cut the bass (lower) frequencies of the audio using a two-pole
1733 shelving filter with a response similar to that of a standard
1734 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1736 The filter accepts the following options:
1740 Give the gain at 0 Hz. Its useful range is about -20
1741 (for a large cut) to +20 (for a large boost).
1742 Beware of clipping when using a positive gain.
1745 Set the filter's central frequency and so can be used
1746 to extend or reduce the frequency range to be boosted or cut.
1747 The default value is @code{100} Hz.
1750 Set method to specify band-width of filter.
1763 Determine how steep is the filter's shelf transition.
1768 Apply a biquad IIR filter with the given coefficients.
1769 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1770 are the numerator and denominator coefficients respectively.
1773 Bauer stereo to binaural transformation, which improves headphone listening of
1774 stereo audio records.
1776 It accepts the following parameters:
1780 Pre-defined crossfeed level.
1784 Default level (fcut=700, feed=50).
1787 Chu Moy circuit (fcut=700, feed=60).
1790 Jan Meier circuit (fcut=650, feed=95).
1795 Cut frequency (in Hz).
1804 Remap input channels to new locations.
1806 It accepts the following parameters:
1808 @item channel_layout
1809 The channel layout of the output stream.
1812 Map channels from input to output. The argument is a '|'-separated list of
1813 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1814 @var{in_channel} form. @var{in_channel} can be either the name of the input
1815 channel (e.g. FL for front left) or its index in the input channel layout.
1816 @var{out_channel} is the name of the output channel or its index in the output
1817 channel layout. If @var{out_channel} is not given then it is implicitly an
1818 index, starting with zero and increasing by one for each mapping.
1821 If no mapping is present, the filter will implicitly map input channels to
1822 output channels, preserving indices.
1824 For example, assuming a 5.1+downmix input MOV file,
1826 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1828 will create an output WAV file tagged as stereo from the downmix channels of
1831 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1833 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:5.1' out.wav
1836 @section channelsplit
1838 Split each channel from an input audio stream into a separate output stream.
1840 It accepts the following parameters:
1842 @item channel_layout
1843 The channel layout of the input stream. The default is "stereo".
1846 For example, assuming a stereo input MP3 file,
1848 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1850 will create an output Matroska file with two audio streams, one containing only
1851 the left channel and the other the right channel.
1853 Split a 5.1 WAV file into per-channel files:
1855 ffmpeg -i in.wav -filter_complex
1856 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1857 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1858 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1863 Add a chorus effect to the audio.
1865 Can make a single vocal sound like a chorus, but can also be applied to instrumentation.
1867 Chorus resembles an echo effect with a short delay, but whereas with echo the delay is
1868 constant, with chorus, it is varied using using sinusoidal or triangular modulation.
1869 The modulation depth defines the range the modulated delay is played before or after
1870 the delay. Hence the delayed sound will sound slower or faster, that is the delayed
1871 sound tuned around the original one, like in a chorus where some vocals are slightly
1874 It accepts the following parameters:
1877 Set input gain. Default is 0.4.
1880 Set output gain. Default is 0.4.
1883 Set delays. A typical delay is around 40ms to 60ms.
1895 @subsection Examples
1901 chorus=0.7:0.9:55:0.4:0.25:2
1907 chorus=0.6:0.9:50|60:0.4|0.32:0.25|0.4:2|1.3
1911 Fuller sounding chorus with three delays:
1913 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
1918 Compress or expand the audio's dynamic range.
1920 It accepts the following parameters:
1926 A list of times in seconds for each channel over which the instantaneous level
1927 of the input signal is averaged to determine its volume. @var{attacks} refers to
1928 increase of volume and @var{decays} refers to decrease of volume. For most
1929 situations, the attack time (response to the audio getting louder) should be
1930 shorter than the decay time, because the human ear is more sensitive to sudden
1931 loud audio than sudden soft audio. A typical value for attack is 0.3 seconds and
1932 a typical value for decay is 0.8 seconds.
1933 If specified number of attacks & decays is lower than number of channels, the last
1934 set attack/decay will be used for all remaining channels.
1937 A list of points for the transfer function, specified in dB relative to the
1938 maximum possible signal amplitude. Each key points list must be defined using
1939 the following syntax: @code{x0/y0|x1/y1|x2/y2|....} or
1940 @code{x0/y0 x1/y1 x2/y2 ....}
1942 The input values must be in strictly increasing order but the transfer function
1943 does not have to be monotonically rising. The point @code{0/0} is assumed but
1944 may be overridden (by @code{0/out-dBn}). Typical values for the transfer
1945 function are @code{-70/-70|-60/-20}.
1948 Set the curve radius in dB for all joints. It defaults to 0.01.
1951 Set the additional gain in dB to be applied at all points on the transfer
1952 function. This allows for easy adjustment of the overall gain.
1956 Set an initial volume, in dB, to be assumed for each channel when filtering
1957 starts. This permits the user to supply a nominal level initially, so that, for
1958 example, a very large gain is not applied to initial signal levels before the
1959 companding has begun to operate. A typical value for audio which is initially
1960 quiet is -90 dB. It defaults to 0.
1963 Set a delay, in seconds. The input audio is analyzed immediately, but audio is
1964 delayed before being fed to the volume adjuster. Specifying a delay
1965 approximately equal to the attack/decay times allows the filter to effectively
1966 operate in predictive rather than reactive mode. It defaults to 0.
1970 @subsection Examples
1974 Make music with both quiet and loud passages suitable for listening to in a
1977 compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
1980 Another example for audio with whisper and explosion parts:
1982 compand=0|0:1|1:-90/-900|-70/-70|-30/-9|0/-3:6:0:0:0
1986 A noise gate for when the noise is at a lower level than the signal:
1988 compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
1992 Here is another noise gate, this time for when the noise is at a higher level
1993 than the signal (making it, in some ways, similar to squelch):
1995 compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
1999 2:1 compression starting at -6dB:
2001 compand=points=-80/-80|-6/-6|0/-3.8|20/3.5
2005 2:1 compression starting at -9dB:
2007 compand=points=-80/-80|-9/-9|0/-5.3|20/2.9
2011 2:1 compression starting at -12dB:
2013 compand=points=-80/-80|-12/-12|0/-6.8|20/1.9
2017 2:1 compression starting at -18dB:
2019 compand=points=-80/-80|-18/-18|0/-9.8|20/0.7
2023 3:1 compression starting at -15dB:
2025 compand=points=-80/-80|-15/-15|0/-10.8|20/-5.2
2031 compand=points=-80/-105|-62/-80|-15.4/-15.4|0/-12|20/-7.6
2037 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
2041 Hard limiter at -6dB:
2043 compand=attacks=0:points=-80/-80|-6/-6|20/-6
2047 Hard limiter at -12dB:
2049 compand=attacks=0:points=-80/-80|-12/-12|20/-12
2053 Hard noise gate at -35 dB:
2055 compand=attacks=0:points=-80/-115|-35.1/-80|-35/-35|20/20
2061 compand=attacks=0:points=-80/-80|-12.4/-12.4|-6/-8|0/-6.8|20/-2.8
2065 @section compensationdelay
2067 Compensation Delay Line is a metric based delay to compensate differing
2068 positions of microphones or speakers.
2070 For example, you have recorded guitar with two microphones placed in
2071 different location. Because the front of sound wave has fixed speed in
2072 normal conditions, the phasing of microphones can vary and depends on
2073 their location and interposition. The best sound mix can be achieved when
2074 these microphones are in phase (synchronized). Note that distance of
2075 ~30 cm between microphones makes one microphone to capture signal in
2076 antiphase to another microphone. That makes the final mix sounding moody.
2077 This filter helps to solve phasing problems by adding different delays
2078 to each microphone track and make them synchronized.
2080 The best result can be reached when you take one track as base and
2081 synchronize other tracks one by one with it.
2082 Remember that synchronization/delay tolerance depends on sample rate, too.
2083 Higher sample rates will give more tolerance.
2085 It accepts the following parameters:
2089 Set millimeters distance. This is compensation distance for fine tuning.
2093 Set cm distance. This is compensation distance for tightening distance setup.
2097 Set meters distance. This is compensation distance for hard distance setup.
2101 Set dry amount. Amount of unprocessed (dry) signal.
2105 Set wet amount. Amount of processed (wet) signal.
2109 Set temperature degree in Celsius. This is the temperature of the environment.
2114 Apply a DC shift to the audio.
2116 This can be useful to remove a DC offset (caused perhaps by a hardware problem
2117 in the recording chain) from the audio. The effect of a DC offset is reduced
2118 headroom and hence volume. The @ref{astats} filter can be used to determine if
2119 a signal has a DC offset.
2123 Set the DC shift, allowed range is [-1, 1]. It indicates the amount to shift
2127 Optional. It should have a value much less than 1 (e.g. 0.05 or 0.02) and is
2128 used to prevent clipping.
2132 Dynamic Audio Normalizer.
2134 This filter applies a certain amount of gain to the input audio in order
2135 to bring its peak magnitude to a target level (e.g. 0 dBFS). However, in
2136 contrast to more "simple" normalization algorithms, the Dynamic Audio
2137 Normalizer *dynamically* re-adjusts the gain factor to the input audio.
2138 This allows for applying extra gain to the "quiet" sections of the audio
2139 while avoiding distortions or clipping the "loud" sections. In other words:
2140 The Dynamic Audio Normalizer will "even out" the volume of quiet and loud
2141 sections, in the sense that the volume of each section is brought to the
2142 same target level. Note, however, that the Dynamic Audio Normalizer achieves
2143 this goal *without* applying "dynamic range compressing". It will retain 100%
2144 of the dynamic range *within* each section of the audio file.
2148 Set the frame length in milliseconds. In range from 10 to 8000 milliseconds.
2149 Default is 500 milliseconds.
2150 The Dynamic Audio Normalizer processes the input audio in small chunks,
2151 referred to as frames. This is required, because a peak magnitude has no
2152 meaning for just a single sample value. Instead, we need to determine the
2153 peak magnitude for a contiguous sequence of sample values. While a "standard"
2154 normalizer would simply use the peak magnitude of the complete file, the
2155 Dynamic Audio Normalizer determines the peak magnitude individually for each
2156 frame. The length of a frame is specified in milliseconds. By default, the
2157 Dynamic Audio Normalizer uses a frame length of 500 milliseconds, which has
2158 been found to give good results with most files.
2159 Note that the exact frame length, in number of samples, will be determined
2160 automatically, based on the sampling rate of the individual input audio file.
2163 Set the Gaussian filter window size. In range from 3 to 301, must be odd
2164 number. Default is 31.
2165 Probably the most important parameter of the Dynamic Audio Normalizer is the
2166 @code{window size} of the Gaussian smoothing filter. The filter's window size
2167 is specified in frames, centered around the current frame. For the sake of
2168 simplicity, this must be an odd number. Consequently, the default value of 31
2169 takes into account the current frame, as well as the 15 preceding frames and
2170 the 15 subsequent frames. Using a larger window results in a stronger
2171 smoothing effect and thus in less gain variation, i.e. slower gain
2172 adaptation. Conversely, using a smaller window results in a weaker smoothing
2173 effect and thus in more gain variation, i.e. faster gain adaptation.
2174 In other words, the more you increase this value, the more the Dynamic Audio
2175 Normalizer will behave like a "traditional" normalization filter. On the
2176 contrary, the more you decrease this value, the more the Dynamic Audio
2177 Normalizer will behave like a dynamic range compressor.
2180 Set the target peak value. This specifies the highest permissible magnitude
2181 level for the normalized audio input. This filter will try to approach the
2182 target peak magnitude as closely as possible, but at the same time it also
2183 makes sure that the normalized signal will never exceed the peak magnitude.
2184 A frame's maximum local gain factor is imposed directly by the target peak
2185 magnitude. The default value is 0.95 and thus leaves a headroom of 5%*.
2186 It is not recommended to go above this value.
2189 Set the maximum gain factor. In range from 1.0 to 100.0. Default is 10.0.
2190 The Dynamic Audio Normalizer determines the maximum possible (local) gain
2191 factor for each input frame, i.e. the maximum gain factor that does not
2192 result in clipping or distortion. The maximum gain factor is determined by
2193 the frame's highest magnitude sample. However, the Dynamic Audio Normalizer
2194 additionally bounds the frame's maximum gain factor by a predetermined
2195 (global) maximum gain factor. This is done in order to avoid excessive gain
2196 factors in "silent" or almost silent frames. By default, the maximum gain
2197 factor is 10.0, For most inputs the default value should be sufficient and
2198 it usually is not recommended to increase this value. Though, for input
2199 with an extremely low overall volume level, it may be necessary to allow even
2200 higher gain factors. Note, however, that the Dynamic Audio Normalizer does
2201 not simply apply a "hard" threshold (i.e. cut off values above the threshold).
2202 Instead, a "sigmoid" threshold function will be applied. This way, the
2203 gain factors will smoothly approach the threshold value, but never exceed that
2207 Set the target RMS. In range from 0.0 to 1.0. Default is 0.0 - disabled.
2208 By default, the Dynamic Audio Normalizer performs "peak" normalization.
2209 This means that the maximum local gain factor for each frame is defined
2210 (only) by the frame's highest magnitude sample. This way, the samples can
2211 be amplified as much as possible without exceeding the maximum signal
2212 level, i.e. without clipping. Optionally, however, the Dynamic Audio
2213 Normalizer can also take into account the frame's root mean square,
2214 abbreviated RMS. In electrical engineering, the RMS is commonly used to
2215 determine the power of a time-varying signal. It is therefore considered
2216 that the RMS is a better approximation of the "perceived loudness" than
2217 just looking at the signal's peak magnitude. Consequently, by adjusting all
2218 frames to a constant RMS value, a uniform "perceived loudness" can be
2219 established. If a target RMS value has been specified, a frame's local gain
2220 factor is defined as the factor that would result in exactly that RMS value.
2221 Note, however, that the maximum local gain factor is still restricted by the
2222 frame's highest magnitude sample, in order to prevent clipping.
2225 Enable channels coupling. By default is enabled.
2226 By default, the Dynamic Audio Normalizer will amplify all channels by the same
2227 amount. This means the same gain factor will be applied to all channels, i.e.
2228 the maximum possible gain factor is determined by the "loudest" channel.
2229 However, in some recordings, it may happen that the volume of the different
2230 channels is uneven, e.g. one channel may be "quieter" than the other one(s).
2231 In this case, this option can be used to disable the channel coupling. This way,
2232 the gain factor will be determined independently for each channel, depending
2233 only on the individual channel's highest magnitude sample. This allows for
2234 harmonizing the volume of the different channels.
2237 Enable DC bias correction. By default is disabled.
2238 An audio signal (in the time domain) is a sequence of sample values.
2239 In the Dynamic Audio Normalizer these sample values are represented in the
2240 -1.0 to 1.0 range, regardless of the original input format. Normally, the
2241 audio signal, or "waveform", should be centered around the zero point.
2242 That means if we calculate the mean value of all samples in a file, or in a
2243 single frame, then the result should be 0.0 or at least very close to that
2244 value. If, however, there is a significant deviation of the mean value from
2245 0.0, in either positive or negative direction, this is referred to as a
2246 DC bias or DC offset. Since a DC bias is clearly undesirable, the Dynamic
2247 Audio Normalizer provides optional DC bias correction.
2248 With DC bias correction enabled, the Dynamic Audio Normalizer will determine
2249 the mean value, or "DC correction" offset, of each input frame and subtract
2250 that value from all of the frame's sample values which ensures those samples
2251 are centered around 0.0 again. Also, in order to avoid "gaps" at the frame
2252 boundaries, the DC correction offset values will be interpolated smoothly
2253 between neighbouring frames.
2256 Enable alternative boundary mode. By default is disabled.
2257 The Dynamic Audio Normalizer takes into account a certain neighbourhood
2258 around each frame. This includes the preceding frames as well as the
2259 subsequent frames. However, for the "boundary" frames, located at the very
2260 beginning and at the very end of the audio file, not all neighbouring
2261 frames are available. In particular, for the first few frames in the audio
2262 file, the preceding frames are not known. And, similarly, for the last few
2263 frames in the audio file, the subsequent frames are not known. Thus, the
2264 question arises which gain factors should be assumed for the missing frames
2265 in the "boundary" region. The Dynamic Audio Normalizer implements two modes
2266 to deal with this situation. The default boundary mode assumes a gain factor
2267 of exactly 1.0 for the missing frames, resulting in a smooth "fade in" and
2268 "fade out" at the beginning and at the end of the input, respectively.
2271 Set the compress factor. In range from 0.0 to 30.0. Default is 0.0.
2272 By default, the Dynamic Audio Normalizer does not apply "traditional"
2273 compression. This means that signal peaks will not be pruned and thus the
2274 full dynamic range will be retained within each local neighbourhood. However,
2275 in some cases it may be desirable to combine the Dynamic Audio Normalizer's
2276 normalization algorithm with a more "traditional" compression.
2277 For this purpose, the Dynamic Audio Normalizer provides an optional compression
2278 (thresholding) function. If (and only if) the compression feature is enabled,
2279 all input frames will be processed by a soft knee thresholding function prior
2280 to the actual normalization process. Put simply, the thresholding function is
2281 going to prune all samples whose magnitude exceeds a certain threshold value.
2282 However, the Dynamic Audio Normalizer does not simply apply a fixed threshold
2283 value. Instead, the threshold value will be adjusted for each individual
2285 In general, smaller parameters result in stronger compression, and vice versa.
2286 Values below 3.0 are not recommended, because audible distortion may appear.
2291 Make audio easier to listen to on headphones.
2293 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
2294 so that when listened to on headphones the stereo image is moved from
2295 inside your head (standard for headphones) to outside and in front of
2296 the listener (standard for speakers).
2302 Apply a two-pole peaking equalisation (EQ) filter. With this
2303 filter, the signal-level at and around a selected frequency can
2304 be increased or decreased, whilst (unlike bandpass and bandreject
2305 filters) that at all other frequencies is unchanged.
2307 In order to produce complex equalisation curves, this filter can
2308 be given several times, each with a different central frequency.
2310 The filter accepts the following options:
2314 Set the filter's central frequency in Hz.
2317 Set method to specify band-width of filter.
2330 Specify the band-width of a filter in width_type units.
2333 Set the required gain or attenuation in dB.
2334 Beware of clipping when using a positive gain.
2337 @subsection Examples
2340 Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
2342 equalizer=f=1000:width_type=h:width=200:g=-10
2346 Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
2348 equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
2352 @section extrastereo
2354 Linearly increases the difference between left and right channels which
2355 adds some sort of "live" effect to playback.
2357 The filter accepts the following option:
2361 Sets the difference coefficient (default: 2.5). 0.0 means mono sound
2362 (average of both channels), with 1.0 sound will be unchanged, with
2363 -1.0 left and right channels will be swapped.
2366 Enable clipping. By default is enabled.
2369 @section firequalizer
2370 Apply FIR Equalization using arbitrary frequency response.
2372 The filter accepts the following option:
2376 Set gain curve equation (in dB). The expression can contain variables:
2379 the evaluated frequency
2383 channel number, set to 0 when multichannels evaluation is disabled
2385 channel id, see libavutil/channel_layout.h, set to the first channel id when
2386 multichannels evaluation is disabled
2390 channel_layout, see libavutil/channel_layout.h
2395 @item gain_interpolate(f)
2396 interpolate gain on frequency f based on gain_entry
2398 This option is also available as command. Default is @code{gain_interpolate(f)}.
2401 Set gain entry for gain_interpolate function. The expression can
2405 store gain entry at frequency f with value g
2407 This option is also available as command.
2410 Set filter delay in seconds. Higher value means more accurate.
2411 Default is @code{0.01}.
2414 Set filter accuracy in Hz. Lower value means more accurate.
2415 Default is @code{5}.
2418 Set window function. Acceptable values are:
2421 rectangular window, useful when gain curve is already smooth
2423 hann window (default)
2429 3-terms continuous 1st derivative nuttall window
2431 minimum 3-terms discontinuous nuttall window
2433 4-terms continuous 1st derivative nuttall window
2435 minimum 4-terms discontinuous nuttall (blackman-nuttall) window
2437 blackman-harris window
2441 If enabled, use fixed number of audio samples. This improves speed when
2442 filtering with large delay. Default is disabled.
2445 Enable multichannels evaluation on gain. Default is disabled.
2448 @subsection Examples
2453 firequalizer=gain='if(lt(f,1000), 0, -INF)'
2456 lowpass at 1000 Hz with gain_entry:
2458 firequalizer=gain_entry='entry(1000,0); entry(1001, -INF)'
2461 custom equalization:
2463 firequalizer=gain_entry='entry(100,0); entry(400, -4); entry(1000, -6); entry(2000, 0)'
2468 firequalizer=delay=0.1:fixed=on
2471 lowpass on left channel, highpass on right channel:
2473 firequalizer=gain='if(eq(chid,1), gain_interpolate(f), if(eq(chid,2), gain_interpolate(1e6+f), 0))'
2474 :gain_entry='entry(1000, 0); entry(1001,-INF); entry(1e6+1000,0)':multi=on
2479 Apply a flanging effect to the audio.
2481 The filter accepts the following options:
2485 Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
2488 Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
2491 Set percentage regeneration (delayed signal feedback). Range from -95 to 95.
2495 Set percentage of delayed signal mixed with original. Range from 0 to 100.
2496 Default value is 71.
2499 Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
2502 Set swept wave shape, can be @var{triangular} or @var{sinusoidal}.
2503 Default value is @var{sinusoidal}.
2506 Set swept wave percentage-shift for multi channel. Range from 0 to 100.
2507 Default value is 25.
2510 Set delay-line interpolation, @var{linear} or @var{quadratic}.
2511 Default is @var{linear}.
2516 Apply a high-pass filter with 3dB point frequency.
2517 The filter can be either single-pole, or double-pole (the default).
2518 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2520 The filter accepts the following options:
2524 Set frequency in Hz. Default is 3000.
2527 Set number of poles. Default is 2.
2530 Set method to specify band-width of filter.
2543 Specify the band-width of a filter in width_type units.
2544 Applies only to double-pole filter.
2545 The default is 0.707q and gives a Butterworth response.
2550 Join multiple input streams into one multi-channel stream.
2552 It accepts the following parameters:
2556 The number of input streams. It defaults to 2.
2558 @item channel_layout
2559 The desired output channel layout. It defaults to stereo.
2562 Map channels from inputs to output. The argument is a '|'-separated list of
2563 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
2564 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
2565 can be either the name of the input channel (e.g. FL for front left) or its
2566 index in the specified input stream. @var{out_channel} is the name of the output
2570 The filter will attempt to guess the mappings when they are not specified
2571 explicitly. It does so by first trying to find an unused matching input channel
2572 and if that fails it picks the first unused input channel.
2574 Join 3 inputs (with properly set channel layouts):
2576 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
2579 Build a 5.1 output from 6 single-channel streams:
2581 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
2582 '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'
2588 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
2590 To enable compilation of this filter you need to configure FFmpeg with
2591 @code{--enable-ladspa}.
2595 Specifies the name of LADSPA plugin library to load. If the environment
2596 variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
2597 each one of the directories specified by the colon separated list in
2598 @env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
2599 this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
2600 @file{/usr/lib/ladspa/}.
2603 Specifies the plugin within the library. Some libraries contain only
2604 one plugin, but others contain many of them. If this is not set filter
2605 will list all available plugins within the specified library.
2608 Set the '|' separated list of controls which are zero or more floating point
2609 values that determine the behavior of the loaded plugin (for example delay,
2611 Controls need to be defined using the following syntax:
2612 c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
2613 @var{valuei} is the value set on the @var{i}-th control.
2614 Alternatively they can be also defined using the following syntax:
2615 @var{value0}|@var{value1}|@var{value2}|..., where
2616 @var{valuei} is the value set on the @var{i}-th control.
2617 If @option{controls} is set to @code{help}, all available controls and
2618 their valid ranges are printed.
2620 @item sample_rate, s
2621 Specify the sample rate, default to 44100. Only used if plugin have
2625 Set the number of samples per channel per each output frame, default
2626 is 1024. Only used if plugin have zero inputs.
2629 Set the minimum duration of the sourced audio. See
2630 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
2631 for the accepted syntax.
2632 Note that the resulting duration may be greater than the specified duration,
2633 as the generated audio is always cut at the end of a complete frame.
2634 If not specified, or the expressed duration is negative, the audio is
2635 supposed to be generated forever.
2636 Only used if plugin have zero inputs.
2640 @subsection Examples
2644 List all available plugins within amp (LADSPA example plugin) library:
2650 List all available controls and their valid ranges for @code{vcf_notch}
2651 plugin from @code{VCF} library:
2653 ladspa=f=vcf:p=vcf_notch:c=help
2657 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
2660 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
2664 Add reverberation to the audio using TAP-plugins
2665 (Tom's Audio Processing plugins):
2667 ladspa=file=tap_reverb:tap_reverb
2671 Generate white noise, with 0.2 amplitude:
2673 ladspa=file=cmt:noise_source_white:c=c0=.2
2677 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
2678 @code{C* Audio Plugin Suite} (CAPS) library:
2680 ladspa=file=caps:Click:c=c1=20'
2684 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
2686 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
2690 Increase volume by 20dB using fast lookahead limiter from Steve Harris
2691 @code{SWH Plugins} collection:
2693 ladspa=fast_lookahead_limiter_1913:fastLookaheadLimiter:20|0|2
2697 Attenuate low frequencies using Multiband EQ from Steve Harris
2698 @code{SWH Plugins} collection:
2700 ladspa=mbeq_1197:mbeq:-24|-24|-24|0|0|0|0|0|0|0|0|0|0|0|0
2704 @subsection Commands
2706 This filter supports the following commands:
2709 Modify the @var{N}-th control value.
2711 If the specified value is not valid, it is ignored and prior one is kept.
2716 Apply a low-pass filter with 3dB point frequency.
2717 The filter can be either single-pole or double-pole (the default).
2718 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
2720 The filter accepts the following options:
2724 Set frequency in Hz. Default is 500.
2727 Set number of poles. Default is 2.
2730 Set method to specify band-width of filter.
2743 Specify the band-width of a filter in width_type units.
2744 Applies only to double-pole filter.
2745 The default is 0.707q and gives a Butterworth response.
2751 Mix channels with specific gain levels. The filter accepts the output
2752 channel layout followed by a set of channels definitions.
2754 This filter is also designed to efficiently remap the channels of an audio
2757 The filter accepts parameters of the form:
2758 "@var{l}|@var{outdef}|@var{outdef}|..."
2762 output channel layout or number of channels
2765 output channel specification, of the form:
2766 "@var{out_name}=[@var{gain}*]@var{in_name}[+[@var{gain}*]@var{in_name}...]"
2769 output channel to define, either a channel name (FL, FR, etc.) or a channel
2770 number (c0, c1, etc.)
2773 multiplicative coefficient for the channel, 1 leaving the volume unchanged
2776 input channel to use, see out_name for details; it is not possible to mix
2777 named and numbered input channels
2780 If the `=' in a channel specification is replaced by `<', then the gains for
2781 that specification will be renormalized so that the total is 1, thus
2782 avoiding clipping noise.
2784 @subsection Mixing examples
2786 For example, if you want to down-mix from stereo to mono, but with a bigger
2787 factor for the left channel:
2789 pan=1c|c0=0.9*c0+0.1*c1
2792 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
2793 7-channels surround:
2795 pan=stereo| FL < FL + 0.5*FC + 0.6*BL + 0.6*SL | FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
2798 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
2799 that should be preferred (see "-ac" option) unless you have very specific
2802 @subsection Remapping examples
2804 The channel remapping will be effective if, and only if:
2807 @item gain coefficients are zeroes or ones,
2808 @item only one input per channel output,
2811 If all these conditions are satisfied, the filter will notify the user ("Pure
2812 channel mapping detected"), and use an optimized and lossless method to do the
2815 For example, if you have a 5.1 source and want a stereo audio stream by
2816 dropping the extra channels:
2818 pan="stereo| c0=FL | c1=FR"
2821 Given the same source, you can also switch front left and front right channels
2822 and keep the input channel layout:
2824 pan="5.1| c0=c1 | c1=c0 | c2=c2 | c3=c3 | c4=c4 | c5=c5"
2827 If the input is a stereo audio stream, you can mute the front left channel (and
2828 still keep the stereo channel layout) with:
2833 Still with a stereo audio stream input, you can copy the right channel in both
2834 front left and right:
2836 pan="stereo| c0=FR | c1=FR"
2841 ReplayGain scanner filter. This filter takes an audio stream as an input and
2842 outputs it unchanged.
2843 At end of filtering it displays @code{track_gain} and @code{track_peak}.
2847 Convert the audio sample format, sample rate and channel layout. It is
2848 not meant to be used directly.
2851 Apply time-stretching and pitch-shifting with librubberband.
2853 The filter accepts the following options:
2857 Set tempo scale factor.
2860 Set pitch scale factor.
2863 Set transients detector.
2864 Possible values are:
2873 Possible values are:
2882 Possible values are:
2889 Set processing window size.
2890 Possible values are:
2899 Possible values are:
2906 Enable formant preservation when shift pitching.
2907 Possible values are:
2915 Possible values are:
2924 Possible values are:
2931 @section sidechaincompress
2933 This filter acts like normal compressor but has the ability to compress
2934 detected signal using second input signal.
2935 It needs two input streams and returns one output stream.
2936 First input stream will be processed depending on second stream signal.
2937 The filtered signal then can be filtered with other filters in later stages of
2938 processing. See @ref{pan} and @ref{amerge} filter.
2940 The filter accepts the following options:
2944 Set input gain. Default is 1. Range is between 0.015625 and 64.
2947 If a signal of second stream raises above this level it will affect the gain
2948 reduction of first stream.
2949 By default is 0.125. Range is between 0.00097563 and 1.
2952 Set a ratio about which the signal is reduced. 1:2 means that if the level
2953 raised 4dB above the threshold, it will be only 2dB above after the reduction.
2954 Default is 2. Range is between 1 and 20.
2957 Amount of milliseconds the signal has to rise above the threshold before gain
2958 reduction starts. Default is 20. Range is between 0.01 and 2000.
2961 Amount of milliseconds the signal has to fall below the threshold before
2962 reduction is decreased again. Default is 250. Range is between 0.01 and 9000.
2965 Set the amount by how much signal will be amplified after processing.
2966 Default is 2. Range is from 1 and 64.
2969 Curve the sharp knee around the threshold to enter gain reduction more softly.
2970 Default is 2.82843. Range is between 1 and 8.
2973 Choose if the @code{average} level between all channels of side-chain stream
2974 or the louder(@code{maximum}) channel of side-chain stream affects the
2975 reduction. Default is @code{average}.
2978 Should the exact signal be taken in case of @code{peak} or an RMS one in case
2979 of @code{rms}. Default is @code{rms} which is mainly smoother.
2982 Set sidechain gain. Default is 1. Range is between 0.015625 and 64.
2985 How much to use compressed signal in output. Default is 1.
2986 Range is between 0 and 1.
2989 @subsection Examples
2993 Full ffmpeg example taking 2 audio inputs, 1st input to be compressed
2994 depending on the signal of 2nd input and later compressed signal to be
2995 merged with 2nd input:
2997 ffmpeg -i main.flac -i sidechain.flac -filter_complex "[1:a]asplit=2[sc][mix];[0:a][sc]sidechaincompress[compr];[compr][mix]amerge"
3001 @section sidechaingate
3003 A sidechain gate acts like a normal (wideband) gate but has the ability to
3004 filter the detected signal before sending it to the gain reduction stage.
3005 Normally a gate uses the full range signal to detect a level above the
3007 For example: If you cut all lower frequencies from your sidechain signal
3008 the gate will decrease the volume of your track only if not enough highs
3009 appear. With this technique you are able to reduce the resonation of a
3010 natural drum or remove "rumbling" of muted strokes from a heavily distorted
3012 It needs two input streams and returns one output stream.
3013 First input stream will be processed depending on second stream signal.
3015 The filter accepts the following options:
3019 Set input level before filtering.
3020 Default is 1. Allowed range is from 0.015625 to 64.
3023 Set the level of gain reduction when the signal is below the threshold.
3024 Default is 0.06125. Allowed range is from 0 to 1.
3027 If a signal rises above this level the gain reduction is released.
3028 Default is 0.125. Allowed range is from 0 to 1.
3031 Set a ratio about which the signal is reduced.
3032 Default is 2. Allowed range is from 1 to 9000.
3035 Amount of milliseconds the signal has to rise above the threshold before gain
3037 Default is 20 milliseconds. Allowed range is from 0.01 to 9000.
3040 Amount of milliseconds the signal has to fall below the threshold before the
3041 reduction is increased again. Default is 250 milliseconds.
3042 Allowed range is from 0.01 to 9000.
3045 Set amount of amplification of signal after processing.
3046 Default is 1. Allowed range is from 1 to 64.
3049 Curve the sharp knee around the threshold to enter gain reduction more softly.
3050 Default is 2.828427125. Allowed range is from 1 to 8.
3053 Choose if exact signal should be taken for detection or an RMS like one.
3054 Default is rms. Can be peak or rms.
3057 Choose if the average level between all channels or the louder channel affects
3059 Default is average. Can be average or maximum.
3062 Set sidechain gain. Default is 1. Range is from 0.015625 to 64.
3065 @section silencedetect
3067 Detect silence in an audio stream.
3069 This filter logs a message when it detects that the input audio volume is less
3070 or equal to a noise tolerance value for a duration greater or equal to the
3071 minimum detected noise duration.
3073 The printed times and duration are expressed in seconds.
3075 The filter accepts the following options:
3079 Set silence duration until notification (default is 2 seconds).
3082 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
3083 specified value) or amplitude ratio. Default is -60dB, or 0.001.
3086 @subsection Examples
3090 Detect 5 seconds of silence with -50dB noise tolerance:
3092 silencedetect=n=-50dB:d=5
3096 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
3097 tolerance in @file{silence.mp3}:
3099 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
3103 @section silenceremove
3105 Remove silence from the beginning, middle or end of the audio.
3107 The filter accepts the following options:
3111 This value is used to indicate if audio should be trimmed at beginning of
3112 the audio. A value of zero indicates no silence should be trimmed from the
3113 beginning. When specifying a non-zero value, it trims audio up until it
3114 finds non-silence. Normally, when trimming silence from beginning of audio
3115 the @var{start_periods} will be @code{1} but it can be increased to higher
3116 values to trim all audio up to specific count of non-silence periods.
3117 Default value is @code{0}.
3119 @item start_duration
3120 Specify the amount of time that non-silence must be detected before it stops
3121 trimming audio. By increasing the duration, bursts of noises can be treated
3122 as silence and trimmed off. Default value is @code{0}.
3124 @item start_threshold
3125 This indicates what sample value should be treated as silence. For digital
3126 audio, a value of @code{0} may be fine but for audio recorded from analog,
3127 you may wish to increase the value to account for background noise.
3128 Can be specified in dB (in case "dB" is appended to the specified value)
3129 or amplitude ratio. Default value is @code{0}.
3132 Set the count for trimming silence from the end of audio.
3133 To remove silence from the middle of a file, specify a @var{stop_periods}
3134 that is negative. This value is then treated as a positive value and is
3135 used to indicate the effect should restart processing as specified by
3136 @var{start_periods}, making it suitable for removing periods of silence
3137 in the middle of the audio.
3138 Default value is @code{0}.
3141 Specify a duration of silence that must exist before audio is not copied any
3142 more. By specifying a higher duration, silence that is wanted can be left in
3144 Default value is @code{0}.
3146 @item stop_threshold
3147 This is the same as @option{start_threshold} but for trimming silence from
3149 Can be specified in dB (in case "dB" is appended to the specified value)
3150 or amplitude ratio. Default value is @code{0}.
3153 This indicate that @var{stop_duration} length of audio should be left intact
3154 at the beginning of each period of silence.
3155 For example, if you want to remove long pauses between words but do not want
3156 to remove the pauses completely. Default value is @code{0}.
3159 Set how is silence detected. Can be @code{rms} or @code{peak}. Second is faster
3160 and works better with digital silence which is exactly 0.
3161 Default value is @code{rms}.
3164 Set ratio used to calculate size of window for detecting silence.
3165 Default value is @code{0.02}. Allowed range is from @code{0} to @code{10}.
3168 @subsection Examples
3172 The following example shows how this filter can be used to start a recording
3173 that does not contain the delay at the start which usually occurs between
3174 pressing the record button and the start of the performance:
3176 silenceremove=1:5:0.02
3180 Trim all silence encountered from begining to end where there is more than 1
3181 second of silence in audio:
3183 silenceremove=0:0:0:-1:1:-90dB
3189 SOFAlizer uses head-related transfer functions (HRTFs) to create virtual
3190 loudspeakers around the user for binaural listening via headphones (audio
3191 formats up to 9 channels supported).
3192 The HRTFs are stored in SOFA files (see @url{http://www.sofacoustics.org/} for a database).
3193 SOFAlizer is developed at the Acoustics Research Institute (ARI) of the
3194 Austrian Academy of Sciences.
3196 To enable compilation of this filter you need to configure FFmpeg with
3197 @code{--enable-netcdf}.
3199 The filter accepts the following options:
3203 Set the SOFA file used for rendering.
3206 Set gain applied to audio. Value is in dB. Default is 0.
3209 Set rotation of virtual loudspeakers in deg. Default is 0.
3212 Set elevation of virtual speakers in deg. Default is 0.
3215 Set distance in meters between loudspeakers and the listener with near-field
3216 HRTFs. Default is 1.
3219 Set processing type. Can be @var{time} or @var{freq}. @var{time} is
3220 processing audio in time domain which is slow.
3221 @var{freq} is processing audio in frequency domain which is fast.
3222 Default is @var{freq}.
3225 Set custom positions of virtual loudspeakers. Syntax for this option is:
3226 <CH> <AZIM> <ELEV>[|<CH> <AZIM> <ELEV>|...].
3227 Each virtual loudspeaker is described with short channel name following with
3228 azimuth and elevation in degreees.
3229 Each virtual loudspeaker description is separated by '|'.
3230 For example to override front left and front right channel positions use:
3231 'speakers=FL 45 15|FR 345 15'.
3232 Descriptions with unrecognised channel names are ignored.
3235 @subsection Examples
3239 Using ClubFritz6 sofa file:
3241 sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=1
3245 Using ClubFritz12 sofa file and bigger radius with small rotation:
3247 sofalizer=sofa=/path/to/ClubFritz12.sofa:type=freq:radius=2:rotation=5
3251 Similar as above but with custom speaker positions for front left, front right, rear left and rear right
3252 and also with custom gain:
3254 "sofalizer=sofa=/path/to/ClubFritz6.sofa:type=freq:radius=2:speakers=FL 45|FR 315|RL 135|RR 225:gain=28"
3258 @section stereotools
3260 This filter has some handy utilities to manage stereo signals, for converting
3261 M/S stereo recordings to L/R signal while having control over the parameters
3262 or spreading the stereo image of master track.
3264 The filter accepts the following options:
3268 Set input level before filtering for both channels. Defaults is 1.
3269 Allowed range is from 0.015625 to 64.
3272 Set output level after filtering for both channels. Defaults is 1.
3273 Allowed range is from 0.015625 to 64.
3276 Set input balance between both channels. Default is 0.
3277 Allowed range is from -1 to 1.
3280 Set output balance between both channels. Default is 0.
3281 Allowed range is from -1 to 1.
3284 Enable softclipping. Results in analog distortion instead of harsh digital 0dB
3285 clipping. Disabled by default.
3288 Mute the left channel. Disabled by default.
3291 Mute the right channel. Disabled by default.
3294 Change the phase of the left channel. Disabled by default.
3297 Change the phase of the right channel. Disabled by default.
3300 Set stereo mode. Available values are:
3304 Left/Right to Left/Right, this is default.
3307 Left/Right to Mid/Side.
3310 Mid/Side to Left/Right.
3313 Left/Right to Left/Left.
3316 Left/Right to Right/Right.
3319 Left/Right to Left + Right.
3322 Left/Right to Right/Left.
3326 Set level of side signal. Default is 1.
3327 Allowed range is from 0.015625 to 64.
3330 Set balance of side signal. Default is 0.
3331 Allowed range is from -1 to 1.
3334 Set level of the middle signal. Default is 1.
3335 Allowed range is from 0.015625 to 64.
3338 Set middle signal pan. Default is 0. Allowed range is from -1 to 1.
3341 Set stereo base between mono and inversed channels. Default is 0.
3342 Allowed range is from -1 to 1.
3345 Set delay in milliseconds how much to delay left from right channel and
3346 vice versa. Default is 0. Allowed range is from -20 to 20.
3349 Set S/C level. Default is 1. Allowed range is from 1 to 100.
3352 Set the stereo phase in degrees. Default is 0. Allowed range is from 0 to 360.
3355 @subsection Examples
3359 Apply karaoke like effect:
3361 stereotools=mlev=0.015625
3365 Convert M/S signal to L/R:
3367 "stereotools=mode=ms>lr"
3371 @section stereowiden
3373 This filter enhance the stereo effect by suppressing signal common to both
3374 channels and by delaying the signal of left into right and vice versa,
3375 thereby widening the stereo effect.
3377 The filter accepts the following options:
3381 Time in milliseconds of the delay of left signal into right and vice versa.
3382 Default is 20 milliseconds.
3385 Amount of gain in delayed signal into right and vice versa. Gives a delay
3386 effect of left signal in right output and vice versa which gives widening
3387 effect. Default is 0.3.
3390 Cross feed of left into right with inverted phase. This helps in suppressing
3391 the mono. If the value is 1 it will cancel all the signal common to both
3392 channels. Default is 0.3.
3395 Set level of input signal of original channel. Default is 0.8.
3400 Boost or cut treble (upper) frequencies of the audio using a two-pole
3401 shelving filter with a response similar to that of a standard
3402 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
3404 The filter accepts the following options:
3408 Give the gain at whichever is the lower of ~22 kHz and the
3409 Nyquist frequency. Its useful range is about -20 (for a large cut)
3410 to +20 (for a large boost). Beware of clipping when using a positive gain.
3413 Set the filter's central frequency and so can be used
3414 to extend or reduce the frequency range to be boosted or cut.
3415 The default value is @code{3000} Hz.
3418 Set method to specify band-width of filter.
3431 Determine how steep is the filter's shelf transition.
3436 Sinusoidal amplitude modulation.
3438 The filter accepts the following options:
3442 Modulation frequency in Hertz. Modulation frequencies in the subharmonic range
3443 (20 Hz or lower) will result in a tremolo effect.
3444 This filter may also be used as a ring modulator by specifying
3445 a modulation frequency higher than 20 Hz.
3446 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3449 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3450 Default value is 0.5.
3455 Sinusoidal phase modulation.
3457 The filter accepts the following options:
3461 Modulation frequency in Hertz.
3462 Range is 0.1 - 20000.0. Default value is 5.0 Hz.
3465 Depth of modulation as a percentage. Range is 0.0 - 1.0.
3466 Default value is 0.5.
3471 Adjust the input audio volume.
3473 It accepts the following parameters:
3477 Set audio volume expression.
3479 Output values are clipped to the maximum value.
3481 The output audio volume is given by the relation:
3483 @var{output_volume} = @var{volume} * @var{input_volume}
3486 The default value for @var{volume} is "1.0".
3489 This parameter represents the mathematical precision.
3491 It determines which input sample formats will be allowed, which affects the
3492 precision of the volume scaling.
3496 8-bit fixed-point; this limits input sample format to U8, S16, and S32.
3498 32-bit floating-point; this limits input sample format to FLT. (default)
3500 64-bit floating-point; this limits input sample format to DBL.
3504 Choose the behaviour on encountering ReplayGain side data in input frames.
3508 Remove ReplayGain side data, ignoring its contents (the default).
3511 Ignore ReplayGain side data, but leave it in the frame.
3514 Prefer the track gain, if present.
3517 Prefer the album gain, if present.
3520 @item replaygain_preamp
3521 Pre-amplification gain in dB to apply to the selected replaygain gain.
3523 Default value for @var{replaygain_preamp} is 0.0.
3526 Set when the volume expression is evaluated.
3528 It accepts the following values:
3531 only evaluate expression once during the filter initialization, or
3532 when the @samp{volume} command is sent
3535 evaluate expression for each incoming frame
3538 Default value is @samp{once}.
3541 The volume expression can contain the following parameters.
3545 frame number (starting at zero)
3548 @item nb_consumed_samples
3549 number of samples consumed by the filter
3551 number of samples in the current frame
3553 original frame position in the file
3559 PTS at start of stream
3561 time at start of stream
3567 last set volume value
3570 Note that when @option{eval} is set to @samp{once} only the
3571 @var{sample_rate} and @var{tb} variables are available, all other
3572 variables will evaluate to NAN.
3574 @subsection Commands
3576 This filter supports the following commands:
3579 Modify the volume expression.
3580 The command accepts the same syntax of the corresponding option.
3582 If the specified expression is not valid, it is kept at its current
3584 @item replaygain_noclip
3585 Prevent clipping by limiting the gain applied.
3587 Default value for @var{replaygain_noclip} is 1.
3591 @subsection Examples
3595 Halve the input audio volume:
3599 volume=volume=-6.0206dB
3602 In all the above example the named key for @option{volume} can be
3603 omitted, for example like in:
3609 Increase input audio power by 6 decibels using fixed-point precision:
3611 volume=volume=6dB:precision=fixed
3615 Fade volume after time 10 with an annihilation period of 5 seconds:
3617 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
3621 @section volumedetect
3623 Detect the volume of the input video.
3625 The filter has no parameters. The input is not modified. Statistics about
3626 the volume will be printed in the log when the input stream end is reached.
3628 In particular it will show the mean volume (root mean square), maximum
3629 volume (on a per-sample basis), and the beginning of a histogram of the
3630 registered volume values (from the maximum value to a cumulated 1/1000 of
3633 All volumes are in decibels relative to the maximum PCM value.
3635 @subsection Examples
3637 Here is an excerpt of the output:
3639 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
3640 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
3641 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
3642 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
3643 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
3644 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
3645 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
3646 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
3647 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
3653 The mean square energy is approximately -27 dB, or 10^-2.7.
3655 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
3657 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
3660 In other words, raising the volume by +4 dB does not cause any clipping,
3661 raising it by +5 dB causes clipping for 6 samples, etc.
3663 @c man end AUDIO FILTERS
3665 @chapter Audio Sources
3666 @c man begin AUDIO SOURCES
3668 Below is a description of the currently available audio sources.
3672 Buffer audio frames, and make them available to the filter chain.
3674 This source is mainly intended for a programmatic use, in particular
3675 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
3677 It accepts the following parameters:
3681 The timebase which will be used for timestamps of submitted frames. It must be
3682 either a floating-point number or in @var{numerator}/@var{denominator} form.
3685 The sample rate of the incoming audio buffers.
3688 The sample format of the incoming audio buffers.
3689 Either a sample format name or its corresponding integer representation from
3690 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
3692 @item channel_layout
3693 The channel layout of the incoming audio buffers.
3694 Either a channel layout name from channel_layout_map in
3695 @file{libavutil/channel_layout.c} or its corresponding integer representation
3696 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
3699 The number of channels of the incoming audio buffers.
3700 If both @var{channels} and @var{channel_layout} are specified, then they
3705 @subsection Examples
3708 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
3711 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
3712 Since the sample format with name "s16p" corresponds to the number
3713 6 and the "stereo" channel layout corresponds to the value 0x3, this is
3716 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
3721 Generate an audio signal specified by an expression.
3723 This source accepts in input one or more expressions (one for each
3724 channel), which are evaluated and used to generate a corresponding
3727 This source accepts the following options:
3731 Set the '|'-separated expressions list for each separate channel. In case the
3732 @option{channel_layout} option is not specified, the selected channel layout
3733 depends on the number of provided expressions. Otherwise the last
3734 specified expression is applied to the remaining output channels.
3736 @item channel_layout, c
3737 Set the channel layout. The number of channels in the specified layout
3738 must be equal to the number of specified expressions.
3741 Set the minimum duration of the sourced audio. See
3742 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
3743 for the accepted syntax.
3744 Note that the resulting duration may be greater than the specified
3745 duration, as the generated audio is always cut at the end of a
3748 If not specified, or the expressed duration is negative, the audio is
3749 supposed to be generated forever.
3752 Set the number of samples per channel per each output frame,
3755 @item sample_rate, s
3756 Specify the sample rate, default to 44100.
3759 Each expression in @var{exprs} can contain the following constants:
3763 number of the evaluated sample, starting from 0
3766 time of the evaluated sample expressed in seconds, starting from 0
3773 @subsection Examples
3783 Generate a sin signal with frequency of 440 Hz, set sample rate to
3786 aevalsrc="sin(440*2*PI*t):s=8000"
3790 Generate a two channels signal, specify the channel layout (Front
3791 Center + Back Center) explicitly:
3793 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
3797 Generate white noise:
3799 aevalsrc="-2+random(0)"
3803 Generate an amplitude modulated signal:
3805 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
3809 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
3811 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
3818 The null audio source, return unprocessed audio frames. It is mainly useful
3819 as a template and to be employed in analysis / debugging tools, or as
3820 the source for filters which ignore the input data (for example the sox
3823 This source accepts the following options:
3827 @item channel_layout, cl
3829 Specifies the channel layout, and can be either an integer or a string
3830 representing a channel layout. The default value of @var{channel_layout}
3833 Check the channel_layout_map definition in
3834 @file{libavutil/channel_layout.c} for the mapping between strings and
3835 channel layout values.
3837 @item sample_rate, r
3838 Specifies the sample rate, and defaults to 44100.
3841 Set the number of samples per requested frames.
3845 @subsection Examples
3849 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
3851 anullsrc=r=48000:cl=4
3855 Do the same operation with a more obvious syntax:
3857 anullsrc=r=48000:cl=mono
3861 All the parameters need to be explicitly defined.
3865 Synthesize a voice utterance using the libflite library.
3867 To enable compilation of this filter you need to configure FFmpeg with
3868 @code{--enable-libflite}.
3870 Note that the flite library is not thread-safe.
3872 The filter accepts the following options:
3877 If set to 1, list the names of the available voices and exit
3878 immediately. Default value is 0.
3881 Set the maximum number of samples per frame. Default value is 512.
3884 Set the filename containing the text to speak.
3887 Set the text to speak.
3890 Set the voice to use for the speech synthesis. Default value is
3891 @code{kal}. See also the @var{list_voices} option.
3894 @subsection Examples
3898 Read from file @file{speech.txt}, and synthesize the text using the
3899 standard flite voice:
3901 flite=textfile=speech.txt
3905 Read the specified text selecting the @code{slt} voice:
3907 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
3911 Input text to ffmpeg:
3913 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
3917 Make @file{ffplay} speak the specified text, using @code{flite} and
3918 the @code{lavfi} device:
3920 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
3924 For more information about libflite, check:
3925 @url{http://www.speech.cs.cmu.edu/flite/}
3929 Generate a noise audio signal.
3931 The filter accepts the following options:
3934 @item sample_rate, r
3935 Specify the sample rate. Default value is 48000 Hz.
3938 Specify the amplitude (0.0 - 1.0) of the generated audio stream. Default value
3942 Specify the duration of the generated audio stream. Not specifying this option
3943 results in noise with an infinite length.
3945 @item color, colour, c
3946 Specify the color of noise. Available noise colors are white, pink, and brown.
3947 Default color is white.
3950 Specify a value used to seed the PRNG.
3953 Set the number of samples per each output frame, default is 1024.
3956 @subsection Examples
3961 Generate 60 seconds of pink noise, with a 44.1 kHz sampling rate and an amplitude of 0.5:
3963 anoisesrc=d=60:c=pink:r=44100:a=0.5
3969 Generate an audio signal made of a sine wave with amplitude 1/8.
3971 The audio signal is bit-exact.
3973 The filter accepts the following options:
3978 Set the carrier frequency. Default is 440 Hz.
3980 @item beep_factor, b
3981 Enable a periodic beep every second with frequency @var{beep_factor} times
3982 the carrier frequency. Default is 0, meaning the beep is disabled.
3984 @item sample_rate, r
3985 Specify the sample rate, default is 44100.
3988 Specify the duration of the generated audio stream.
3990 @item samples_per_frame
3991 Set the number of samples per output frame.
3993 The expression can contain the following constants:
3997 The (sequential) number of the output audio frame, starting from 0.
4000 The PTS (Presentation TimeStamp) of the output audio frame,
4001 expressed in @var{TB} units.
4004 The PTS of the output audio frame, expressed in seconds.
4007 The timebase of the output audio frames.
4010 Default is @code{1024}.
4013 @subsection Examples
4018 Generate a simple 440 Hz sine wave:
4024 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
4028 sine=frequency=220:beep_factor=4:duration=5
4032 Generate a 1 kHz sine wave following @code{1602,1601,1602,1601,1602} NTSC
4035 sine=1000:samples_per_frame='st(0,mod(n,5)); 1602-not(not(eq(ld(0),1)+eq(ld(0),3)))'
4039 @c man end AUDIO SOURCES
4041 @chapter Audio Sinks
4042 @c man begin AUDIO SINKS
4044 Below is a description of the currently available audio sinks.
4046 @section abuffersink
4048 Buffer audio frames, and make them available to the end of filter chain.
4050 This sink is mainly intended for programmatic use, in particular
4051 through the interface defined in @file{libavfilter/buffersink.h}
4052 or the options system.
4054 It accepts a pointer to an AVABufferSinkContext structure, which
4055 defines the incoming buffers' formats, to be passed as the opaque
4056 parameter to @code{avfilter_init_filter} for initialization.
4059 Null audio sink; do absolutely nothing with the input audio. It is
4060 mainly useful as a template and for use in analysis / debugging
4063 @c man end AUDIO SINKS
4065 @chapter Video Filters
4066 @c man begin VIDEO FILTERS
4068 When you configure your FFmpeg build, you can disable any of the
4069 existing filters using @code{--disable-filters}.
4070 The configure output will show the video filters included in your
4073 Below is a description of the currently available video filters.
4075 @section alphaextract
4077 Extract the alpha component from the input as a grayscale video. This
4078 is especially useful with the @var{alphamerge} filter.
4082 Add or replace the alpha component of the primary input with the
4083 grayscale value of a second input. This is intended for use with
4084 @var{alphaextract} to allow the transmission or storage of frame
4085 sequences that have alpha in a format that doesn't support an alpha
4088 For example, to reconstruct full frames from a normal YUV-encoded video
4089 and a separate video created with @var{alphaextract}, you might use:
4091 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
4094 Since this filter is designed for reconstruction, it operates on frame
4095 sequences without considering timestamps, and terminates when either
4096 input reaches end of stream. This will cause problems if your encoding
4097 pipeline drops frames. If you're trying to apply an image as an
4098 overlay to a video stream, consider the @var{overlay} filter instead.
4102 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
4103 and libavformat to work. On the other hand, it is limited to ASS (Advanced
4104 Substation Alpha) subtitles files.
4106 This filter accepts the following option in addition to the common options from
4107 the @ref{subtitles} filter:
4111 Set the shaping engine
4113 Available values are:
4116 The default libass shaping engine, which is the best available.
4118 Fast, font-agnostic shaper that can do only substitutions
4120 Slower shaper using OpenType for substitutions and positioning
4123 The default is @code{auto}.
4127 Apply an Adaptive Temporal Averaging Denoiser to the video input.
4129 The filter accepts the following options:
4133 Set threshold A for 1st plane. Default is 0.02.
4134 Valid range is 0 to 0.3.
4137 Set threshold B for 1st plane. Default is 0.04.
4138 Valid range is 0 to 5.
4141 Set threshold A for 2nd plane. Default is 0.02.
4142 Valid range is 0 to 0.3.
4145 Set threshold B for 2nd plane. Default is 0.04.
4146 Valid range is 0 to 5.
4149 Set threshold A for 3rd plane. Default is 0.02.
4150 Valid range is 0 to 0.3.
4153 Set threshold B for 3rd plane. Default is 0.04.
4154 Valid range is 0 to 5.
4156 Threshold A is designed to react on abrupt changes in the input signal and
4157 threshold B is designed to react on continuous changes in the input signal.
4160 Set number of frames filter will use for averaging. Default is 33. Must be odd
4161 number in range [5, 129].
4166 Compute the bounding box for the non-black pixels in the input frame
4169 This filter computes the bounding box containing all the pixels with a
4170 luminance value greater than the minimum allowed value.
4171 The parameters describing the bounding box are printed on the filter
4174 The filter accepts the following option:
4178 Set the minimal luminance value. Default is @code{16}.
4181 @section blackdetect
4183 Detect video intervals that are (almost) completely black. Can be
4184 useful to detect chapter transitions, commercials, or invalid
4185 recordings. Output lines contains the time for the start, end and
4186 duration of the detected black interval expressed in seconds.
4188 In order to display the output lines, you need to set the loglevel at
4189 least to the AV_LOG_INFO value.
4191 The filter accepts the following options:
4194 @item black_min_duration, d
4195 Set the minimum detected black duration expressed in seconds. It must
4196 be a non-negative floating point number.
4198 Default value is 2.0.
4200 @item picture_black_ratio_th, pic_th
4201 Set the threshold for considering a picture "black".
4202 Express the minimum value for the ratio:
4204 @var{nb_black_pixels} / @var{nb_pixels}
4207 for which a picture is considered black.
4208 Default value is 0.98.
4210 @item pixel_black_th, pix_th
4211 Set the threshold for considering a pixel "black".
4213 The threshold expresses the maximum pixel luminance value for which a
4214 pixel is considered "black". The provided value is scaled according to
4215 the following equation:
4217 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
4220 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
4221 the input video format, the range is [0-255] for YUV full-range
4222 formats and [16-235] for YUV non full-range formats.
4224 Default value is 0.10.
4227 The following example sets the maximum pixel threshold to the minimum
4228 value, and detects only black intervals of 2 or more seconds:
4230 blackdetect=d=2:pix_th=0.00
4235 Detect frames that are (almost) completely black. Can be useful to
4236 detect chapter transitions or commercials. Output lines consist of
4237 the frame number of the detected frame, the percentage of blackness,
4238 the position in the file if known or -1 and the timestamp in seconds.
4240 In order to display the output lines, you need to set the loglevel at
4241 least to the AV_LOG_INFO value.
4243 It accepts the following parameters:
4248 The percentage of the pixels that have to be below the threshold; it defaults to
4251 @item threshold, thresh
4252 The threshold below which a pixel value is considered black; it defaults to
4257 @section blend, tblend
4259 Blend two video frames into each other.
4261 The @code{blend} filter takes two input streams and outputs one
4262 stream, the first input is the "top" layer and second input is
4263 "bottom" layer. Output terminates when shortest input terminates.
4265 The @code{tblend} (time blend) filter takes two consecutive frames
4266 from one single stream, and outputs the result obtained by blending
4267 the new frame on top of the old frame.
4269 A description of the accepted options follows.
4277 Set blend mode for specific pixel component or all pixel components in case
4278 of @var{all_mode}. Default value is @code{normal}.
4280 Available values for component modes are:
4321 Set blend opacity for specific pixel component or all pixel components in case
4322 of @var{all_opacity}. Only used in combination with pixel component blend modes.
4329 Set blend expression for specific pixel component or all pixel components in case
4330 of @var{all_expr}. Note that related mode options will be ignored if those are set.
4332 The expressions can use the following variables:
4336 The sequential number of the filtered frame, starting from @code{0}.
4340 the coordinates of the current sample
4344 the width and height of currently filtered plane
4348 Width and height scale depending on the currently filtered plane. It is the
4349 ratio between the corresponding luma plane number of pixels and the current
4350 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
4351 @code{0.5,0.5} for chroma planes.
4354 Time of the current frame, expressed in seconds.
4357 Value of pixel component at current location for first video frame (top layer).
4360 Value of pixel component at current location for second video frame (bottom layer).
4364 Force termination when the shortest input terminates. Default is
4365 @code{0}. This option is only defined for the @code{blend} filter.
4368 Continue applying the last bottom frame after the end of the stream. A value of
4369 @code{0} disable the filter after the last frame of the bottom layer is reached.
4370 Default is @code{1}. This option is only defined for the @code{blend} filter.
4373 @subsection Examples
4377 Apply transition from bottom layer to top layer in first 10 seconds:
4379 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
4383 Apply 1x1 checkerboard effect:
4385 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
4389 Apply uncover left effect:
4391 blend=all_expr='if(gte(N*SW+X,W),A,B)'
4395 Apply uncover down effect:
4397 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
4401 Apply uncover up-left effect:
4403 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
4407 Split diagonally video and shows top and bottom layer on each side:
4409 blend=all_expr=if(gt(X,Y*(W/H)),A,B)
4413 Display differences between the current and the previous frame:
4415 tblend=all_mode=difference128
4421 Deinterlace the input video ("bwdif" stands for "Bob Weaver
4422 Deinterlacing Filter").
4424 Motion adaptive deinterlacing based on yadif with the use of w3fdif and cubic
4425 interpolation algorithms.
4426 It accepts the following parameters:
4430 The interlacing mode to adopt. It accepts one of the following values:
4434 Output one frame for each frame.
4436 Output one frame for each field.
4439 The default value is @code{send_field}.
4442 The picture field parity assumed for the input interlaced video. It accepts one
4443 of the following values:
4447 Assume the top field is first.
4449 Assume the bottom field is first.
4451 Enable automatic detection of field parity.
4454 The default value is @code{auto}.
4455 If the interlacing is unknown or the decoder does not export this information,
4456 top field first will be assumed.
4459 Specify which frames to deinterlace. Accept one of the following
4464 Deinterlace all frames.
4466 Only deinterlace frames marked as interlaced.
4469 The default value is @code{all}.
4474 Apply a boxblur algorithm to the input video.
4476 It accepts the following parameters:
4480 @item luma_radius, lr
4481 @item luma_power, lp
4482 @item chroma_radius, cr
4483 @item chroma_power, cp
4484 @item alpha_radius, ar
4485 @item alpha_power, ap
4489 A description of the accepted options follows.
4492 @item luma_radius, lr
4493 @item chroma_radius, cr
4494 @item alpha_radius, ar
4495 Set an expression for the box radius in pixels used for blurring the
4496 corresponding input plane.
4498 The radius value must be a non-negative number, and must not be
4499 greater than the value of the expression @code{min(w,h)/2} for the
4500 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
4503 Default value for @option{luma_radius} is "2". If not specified,
4504 @option{chroma_radius} and @option{alpha_radius} default to the
4505 corresponding value set for @option{luma_radius}.
4507 The expressions can contain the following constants:
4511 The input width and height in pixels.
4515 The input chroma image width and height in pixels.
4519 The horizontal and vertical chroma subsample values. For example, for the
4520 pixel format "yuv422p", @var{hsub} is 2 and @var{vsub} is 1.
4523 @item luma_power, lp
4524 @item chroma_power, cp
4525 @item alpha_power, ap
4526 Specify how many times the boxblur filter is applied to the
4527 corresponding plane.
4529 Default value for @option{luma_power} is 2. If not specified,
4530 @option{chroma_power} and @option{alpha_power} default to the
4531 corresponding value set for @option{luma_power}.
4533 A value of 0 will disable the effect.
4536 @subsection Examples
4540 Apply a boxblur filter with the luma, chroma, and alpha radii
4543 boxblur=luma_radius=2:luma_power=1
4548 Set the luma radius to 2, and alpha and chroma radius to 0:
4550 boxblur=2:1:cr=0:ar=0
4554 Set the luma and chroma radii to a fraction of the video dimension:
4556 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
4561 YUV colorspace color/chroma keying.
4563 The filter accepts the following options:
4567 The color which will be replaced with transparency.
4570 Similarity percentage with the key color.
4572 0.01 matches only the exact key color, while 1.0 matches everything.
4577 0.0 makes pixels either fully transparent, or not transparent at all.
4579 Higher values result in semi-transparent pixels, with a higher transparency
4580 the more similar the pixels color is to the key color.
4583 Signals that the color passed is already in YUV instead of RGB.
4585 Litteral colors like "green" or "red" don't make sense with this enabled anymore.
4586 This can be used to pass exact YUV values as hexadecimal numbers.
4589 @subsection Examples
4593 Make every green pixel in the input image transparent:
4595 ffmpeg -i input.png -vf chromakey=green out.png
4599 Overlay a greenscreen-video on top of a static black background.
4601 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
4607 Display CIE color diagram with pixels overlaid onto it.
4609 The filter acccepts the following options:
4624 @item uhdtv, rec2020
4637 Set what gamuts to draw.
4639 See @code{system} option for avaiable values.
4642 Set ciescope size, by default set to 512.
4645 Set intensity used to map input pixel values to CIE diagram.
4648 Set contrast used to draw tongue colors that are out of active color system gamut.
4651 Correct gamma displayed on scope, by default enabled.
4654 Show white point on CIE diagram, by default disabled.
4657 Set input gamma. Used only with XYZ input color space.
4662 Visualize information exported by some codecs.
4664 Some codecs can export information through frames using side-data or other
4665 means. For example, some MPEG based codecs export motion vectors through the
4666 @var{export_mvs} flag in the codec @option{flags2} option.
4668 The filter accepts the following option:
4672 Set motion vectors to visualize.
4674 Available flags for @var{mv} are:
4678 forward predicted MVs of P-frames
4680 forward predicted MVs of B-frames
4682 backward predicted MVs of B-frames
4686 Display quantization parameters using the chroma planes
4689 @subsection Examples
4693 Visualizes multi-directionals MVs from P and B-Frames using @command{ffplay}:
4695 ffplay -flags2 +export_mvs input.mpg -vf codecview=mv=pf+bf+bb
4699 @section colorbalance
4700 Modify intensity of primary colors (red, green and blue) of input frames.
4702 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
4703 regions for the red-cyan, green-magenta or blue-yellow balance.
4705 A positive adjustment value shifts the balance towards the primary color, a negative
4706 value towards the complementary color.
4708 The filter accepts the following options:
4714 Adjust red, green and blue shadows (darkest pixels).
4719 Adjust red, green and blue midtones (medium pixels).
4724 Adjust red, green and blue highlights (brightest pixels).
4726 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
4729 @subsection Examples
4733 Add red color cast to shadows:
4740 RGB colorspace color keying.
4742 The filter accepts the following options:
4746 The color which will be replaced with transparency.
4749 Similarity percentage with the key color.
4751 0.01 matches only the exact key color, while 1.0 matches everything.
4756 0.0 makes pixels either fully transparent, or not transparent at all.
4758 Higher values result in semi-transparent pixels, with a higher transparency
4759 the more similar the pixels color is to the key color.
4762 @subsection Examples
4766 Make every green pixel in the input image transparent:
4768 ffmpeg -i input.png -vf colorkey=green out.png
4772 Overlay a greenscreen-video on top of a static background image.
4774 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
4778 @section colorlevels
4780 Adjust video input frames using levels.
4782 The filter accepts the following options:
4789 Adjust red, green, blue and alpha input black point.
4790 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
4796 Adjust red, green, blue and alpha input white point.
4797 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{1}.
4799 Input levels are used to lighten highlights (bright tones), darken shadows
4800 (dark tones), change the balance of bright and dark tones.
4806 Adjust red, green, blue and alpha output black point.
4807 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{0}.
4813 Adjust red, green, blue and alpha output white point.
4814 Allowed ranges for options are @code{[0, 1.0]}. Defaults are @code{1}.
4816 Output levels allows manual selection of a constrained output level range.
4819 @subsection Examples
4823 Make video output darker:
4825 colorlevels=rimin=0.058:gimin=0.058:bimin=0.058
4831 colorlevels=rimin=0.039:gimin=0.039:bimin=0.039:rimax=0.96:gimax=0.96:bimax=0.96
4835 Make video output lighter:
4837 colorlevels=rimax=0.902:gimax=0.902:bimax=0.902
4841 Increase brightness:
4843 colorlevels=romin=0.5:gomin=0.5:bomin=0.5
4847 @section colorchannelmixer
4849 Adjust video input frames by re-mixing color channels.
4851 This filter modifies a color channel by adding the values associated to
4852 the other channels of the same pixels. For example if the value to
4853 modify is red, the output value will be:
4855 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
4858 The filter accepts the following options:
4865 Adjust contribution of input red, green, blue and alpha channels for output red channel.
4866 Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
4872 Adjust contribution of input red, green, blue and alpha channels for output green channel.
4873 Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
4879 Adjust contribution of input red, green, blue and alpha channels for output blue channel.
4880 Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
4886 Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
4887 Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
4889 Allowed ranges for options are @code{[-2.0, 2.0]}.
4892 @subsection Examples
4896 Convert source to grayscale:
4898 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
4901 Simulate sepia tones:
4903 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
4907 @section colormatrix
4909 Convert color matrix.
4911 The filter accepts the following options:
4916 Specify the source and destination color matrix. Both values must be
4919 The accepted values are:
4935 For example to convert from BT.601 to SMPTE-240M, use the command:
4937 colormatrix=bt601:smpte240m
4942 Convert colorspace, transfer characteristics or color primaries.
4944 The filter accepts the following options:
4948 Specify all color properties at once.
4950 The accepted values are:
4979 Specify output colorspace.
4981 The accepted values are:
4990 BT.470BG or BT.601-6 625
4993 SMPTE-170M or BT.601-6 525
4999 BT.2020 with non-constant luminance
5004 Specify output transfer characteristics.
5006 The accepted values are:
5012 Constant gamma of 2.2
5015 Constant gamma of 2.8
5018 SMPTE-170M, BT.601-6 625 or BT.601-6 525
5024 BT.2020 for 10-bits content
5027 BT.2020 for 12-bits content
5032 Specify output color primaries.
5034 The accepted values are:
5043 BT.470BG or BT.601-6 625
5046 SMPTE-170M or BT.601-6 525
5057 Specify output color range.
5059 The accepted values are:
5062 MPEG (restricted) range
5070 Specify output color format.
5072 The accepted values are:
5075 YUV 4:2:0 planar 8-bits
5078 YUV 4:2:0 planar 10-bits
5081 YUV 4:2:0 planar 12-bits
5084 YUV 4:2:2 planar 8-bits
5087 YUV 4:2:2 planar 10-bits
5090 YUV 4:2:2 planar 12-bits
5093 YUV 4:4:4 planar 8-bits
5096 YUV 4:4:4 planar 10-bits
5099 YUV 4:4:4 planar 12-bits
5104 Do a fast conversion, which skips gamma/primary correction. This will take
5105 significantly less CPU, but will be mathematically incorrect. To get output
5106 compatible with that produced by the colormatrix filter, use fast=1.
5109 The filter converts the transfer characteristics, color space and color
5110 primaries to the specified user values. The output value, if not specified,
5111 is set to a default value based on the "all" property. If that property is
5112 also not specified, the filter will log an error. The output color range and
5113 format default to the same value as the input color range and format. The
5114 input transfer characteristics, color space, color primaries and color range
5115 should be set on the input data. If any of these are missing, the filter will
5116 log an error and no conversion will take place.
5118 For example to convert the input to SMPTE-240M, use the command:
5120 colorspace=smpte240m
5123 @section convolution
5125 Apply convolution 3x3 or 5x5 filter.
5127 The filter accepts the following options:
5134 Set matrix for each plane.
5135 Matrix is sequence of 9 or 25 signed integers.
5141 Set multiplier for calculated value for each plane.
5147 Set bias for each plane. This value is added to the result of the multiplication.
5148 Useful for making the overall image brighter or darker. Default is 0.0.
5151 @subsection Examples
5157 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"
5163 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"
5169 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"
5175 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"
5181 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"
5187 Copy the input source unchanged to the output. This is mainly useful for
5192 Video filtering on GPU using Apple's CoreImage API on OSX.
5194 Hardware acceleration is based on an OpenGL context. Usually, this means it is
5195 processed by video hardware. However, software-based OpenGL implementations
5196 exist which means there is no guarantee for hardware processing. It depends on
5199 There are many filters and image generators provided by Apple that come with a
5200 large variety of options. The filter has to be referenced by its name along
5203 The coreimage filter accepts the following options:
5206 List all available filters and generators along with all their respective
5207 options as well as possible minimum and maximum values along with the default
5214 Specify all filters by their respective name and options.
5215 Use @var{list_filters} to determine all valid filter names and options.
5216 Numerical options are specified by a float value and are automatically clamped
5217 to their respective value range. Vector and color options have to be specified
5218 by a list of space separated float values. Character escaping has to be done.
5219 A special option name @code{default} is available to use default options for a
5222 It is required to specify either @code{default} or at least one of the filter options.
5223 All omitted options are used with their default values.
5224 The syntax of the filter string is as follows:
5226 filter=<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...][#<NAME>@@<OPTION>=<VALUE>[@@<OPTION>=<VALUE>][@@...]][#...]
5230 Specify a rectangle where the output of the filter chain is copied into the
5231 input image. It is given by a list of space separated float values:
5233 output_rect=x\ y\ width\ height
5235 If not given, the output rectangle equals the dimensions of the input image.
5236 The output rectangle is automatically cropped at the borders of the input
5237 image. Negative values are valid for each component.
5239 output_rect=25\ 25\ 100\ 100
5243 Several filters can be chained for successive processing without GPU-HOST
5244 transfers allowing for fast processing of complex filter chains.
5245 Currently, only filters with zero (generators) or exactly one (filters) input
5246 image and one output image are supported. Also, transition filters are not yet
5249 Some filters generate output images with additional padding depending on the
5250 respective filter kernel. The padding is automatically removed to ensure the
5251 filter output has the same size as the input image.
5253 For image generators, the size of the output image is determined by the
5254 previous output image of the filter chain or the input image of the whole
5255 filterchain, respectively. The generators do not use the pixel information of
5256 this image to generate their output. However, the generated output is
5257 blended onto this image, resulting in partial or complete coverage of the
5260 The @ref{coreimagesrc} video source can be used for generating input images
5261 which are directly fed into the filter chain. By using it, providing input
5262 images by another video source or an input video is not required.
5264 @subsection Examples
5269 List all filters available:
5271 coreimage=list_filters=true
5275 Use the CIBoxBlur filter with default options to blur an image:
5277 coreimage=filter=CIBoxBlur@@default
5281 Use a filter chain with CISepiaTone at default values and CIVignetteEffect with
5282 its center at 100x100 and a radius of 50 pixels:
5284 coreimage=filter=CIBoxBlur@@default#CIVignetteEffect@@inputCenter=100\ 100@@inputRadius=50
5288 Use nullsrc and CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
5289 given as complete and escaped command-line for Apple's standard bash shell:
5291 ffmpeg -f lavfi -i nullsrc=s=100x100,coreimage=filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
5297 Crop the input video to given dimensions.
5299 It accepts the following parameters:
5303 The width of the output video. It defaults to @code{iw}.
5304 This expression is evaluated only once during the filter
5305 configuration, or when the @samp{w} or @samp{out_w} command is sent.
5308 The height of the output video. It defaults to @code{ih}.
5309 This expression is evaluated only once during the filter
5310 configuration, or when the @samp{h} or @samp{out_h} command is sent.
5313 The horizontal position, in the input video, of the left edge of the output
5314 video. It defaults to @code{(in_w-out_w)/2}.
5315 This expression is evaluated per-frame.
5318 The vertical position, in the input video, of the top edge of the output video.
5319 It defaults to @code{(in_h-out_h)/2}.
5320 This expression is evaluated per-frame.
5323 If set to 1 will force the output display aspect ratio
5324 to be the same of the input, by changing the output sample aspect
5325 ratio. It defaults to 0.
5328 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
5329 expressions containing the following constants:
5334 The computed values for @var{x} and @var{y}. They are evaluated for
5339 The input width and height.
5343 These are the same as @var{in_w} and @var{in_h}.
5347 The output (cropped) width and height.
5351 These are the same as @var{out_w} and @var{out_h}.
5354 same as @var{iw} / @var{ih}
5357 input sample aspect ratio
5360 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
5364 horizontal and vertical chroma subsample values. For example for the
5365 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
5368 The number of the input frame, starting from 0.
5371 the position in the file of the input frame, NAN if unknown
5374 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
5378 The expression for @var{out_w} may depend on the value of @var{out_h},
5379 and the expression for @var{out_h} may depend on @var{out_w}, but they
5380 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
5381 evaluated after @var{out_w} and @var{out_h}.
5383 The @var{x} and @var{y} parameters specify the expressions for the
5384 position of the top-left corner of the output (non-cropped) area. They
5385 are evaluated for each frame. If the evaluated value is not valid, it
5386 is approximated to the nearest valid value.
5388 The expression for @var{x} may depend on @var{y}, and the expression
5389 for @var{y} may depend on @var{x}.
5391 @subsection Examples
5395 Crop area with size 100x100 at position (12,34).
5400 Using named options, the example above becomes:
5402 crop=w=100:h=100:x=12:y=34
5406 Crop the central input area with size 100x100:
5412 Crop the central input area with size 2/3 of the input video:
5414 crop=2/3*in_w:2/3*in_h
5418 Crop the input video central square:
5425 Delimit the rectangle with the top-left corner placed at position
5426 100:100 and the right-bottom corner corresponding to the right-bottom
5427 corner of the input image.
5429 crop=in_w-100:in_h-100:100:100
5433 Crop 10 pixels from the left and right borders, and 20 pixels from
5434 the top and bottom borders
5436 crop=in_w-2*10:in_h-2*20
5440 Keep only the bottom right quarter of the input image:
5442 crop=in_w/2:in_h/2:in_w/2:in_h/2
5446 Crop height for getting Greek harmony:
5448 crop=in_w:1/PHI*in_w
5452 Apply trembling effect:
5454 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)
5458 Apply erratic camera effect depending on timestamp:
5460 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)"
5464 Set x depending on the value of y:
5466 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
5470 @subsection Commands
5472 This filter supports the following commands:
5478 Set width/height of the output video and the horizontal/vertical position
5480 The command accepts the same syntax of the corresponding option.
5482 If the specified expression is not valid, it is kept at its current
5488 Auto-detect the crop size.
5490 It calculates the necessary cropping parameters and prints the
5491 recommended parameters via the logging system. The detected dimensions
5492 correspond to the non-black area of the input video.
5494 It accepts the following parameters:
5499 Set higher black value threshold, which can be optionally specified
5500 from nothing (0) to everything (255 for 8bit based formats). An intensity
5501 value greater to the set value is considered non-black. It defaults to 24.
5502 You can also specify a value between 0.0 and 1.0 which will be scaled depending
5503 on the bitdepth of the pixel format.
5506 The value which the width/height should be divisible by. It defaults to
5507 16. The offset is automatically adjusted to center the video. Use 2 to
5508 get only even dimensions (needed for 4:2:2 video). 16 is best when
5509 encoding to most video codecs.
5511 @item reset_count, reset
5512 Set the counter that determines after how many frames cropdetect will
5513 reset the previously detected largest video area and start over to
5514 detect the current optimal crop area. Default value is 0.
5516 This can be useful when channel logos distort the video area. 0
5517 indicates 'never reset', and returns the largest area encountered during
5524 Apply color adjustments using curves.
5526 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
5527 component (red, green and blue) has its values defined by @var{N} key points
5528 tied from each other using a smooth curve. The x-axis represents the pixel
5529 values from the input frame, and the y-axis the new pixel values to be set for
5532 By default, a component curve is defined by the two points @var{(0;0)} and
5533 @var{(1;1)}. This creates a straight line where each original pixel value is
5534 "adjusted" to its own value, which means no change to the image.
5536 The filter allows you to redefine these two points and add some more. A new
5537 curve (using a natural cubic spline interpolation) will be define to pass
5538 smoothly through all these new coordinates. The new defined points needs to be
5539 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
5540 be in the @var{[0;1]} interval. If the computed curves happened to go outside
5541 the vector spaces, the values will be clipped accordingly.
5543 If there is no key point defined in @code{x=0}, the filter will automatically
5544 insert a @var{(0;0)} point. In the same way, if there is no key point defined
5545 in @code{x=1}, the filter will automatically insert a @var{(1;1)} point.
5547 The filter accepts the following options:
5551 Select one of the available color presets. This option can be used in addition
5552 to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
5553 options takes priority on the preset values.
5554 Available presets are:
5557 @item color_negative
5560 @item increase_contrast
5562 @item linear_contrast
5563 @item medium_contrast
5565 @item strong_contrast
5568 Default is @code{none}.
5570 Set the master key points. These points will define a second pass mapping. It
5571 is sometimes called a "luminance" or "value" mapping. It can be used with
5572 @option{r}, @option{g}, @option{b} or @option{all} since it acts like a
5573 post-processing LUT.
5575 Set the key points for the red component.
5577 Set the key points for the green component.
5579 Set the key points for the blue component.
5581 Set the key points for all components (not including master).
5582 Can be used in addition to the other key points component
5583 options. In this case, the unset component(s) will fallback on this
5584 @option{all} setting.
5586 Specify a Photoshop curves file (@code{.acv}) to import the settings from.
5589 To avoid some filtergraph syntax conflicts, each key points list need to be
5590 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
5592 @subsection Examples
5596 Increase slightly the middle level of blue:
5598 curves=blue='0.5/0.58'
5604 curves=r='0/0.11 .42/.51 1/0.95':g='0.50/0.48':b='0/0.22 .49/.44 1/0.8'
5606 Here we obtain the following coordinates for each components:
5609 @code{(0;0.11) (0.42;0.51) (1;0.95)}
5611 @code{(0;0) (0.50;0.48) (1;1)}
5613 @code{(0;0.22) (0.49;0.44) (1;0.80)}
5617 The previous example can also be achieved with the associated built-in preset:
5619 curves=preset=vintage
5629 Use a Photoshop preset and redefine the points of the green component:
5631 curves=psfile='MyCurvesPresets/purple.acv':green='0.45/0.53'
5637 Video data analysis filter.
5639 This filter shows hexadecimal pixel values of part of video.
5641 The filter accepts the following options:
5645 Set output video size.
5648 Set x offset from where to pick pixels.
5651 Set y offset from where to pick pixels.
5654 Set scope mode, can be one of the following:
5657 Draw hexadecimal pixel values with white color on black background.
5660 Draw hexadecimal pixel values with input video pixel color on black
5664 Draw hexadecimal pixel values on color background picked from input video,
5665 the text color is picked in such way so its always visible.
5669 Draw rows and columns numbers on left and top of video.
5674 Denoise frames using 2D DCT (frequency domain filtering).
5676 This filter is not designed for real time.
5678 The filter accepts the following options:
5682 Set the noise sigma constant.
5684 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
5685 coefficient (absolute value) below this threshold with be dropped.
5687 If you need a more advanced filtering, see @option{expr}.
5689 Default is @code{0}.
5692 Set number overlapping pixels for each block. Since the filter can be slow, you
5693 may want to reduce this value, at the cost of a less effective filter and the
5694 risk of various artefacts.
5696 If the overlapping value doesn't permit processing the whole input width or
5697 height, a warning will be displayed and according borders won't be denoised.
5699 Default value is @var{blocksize}-1, which is the best possible setting.
5702 Set the coefficient factor expression.
5704 For each coefficient of a DCT block, this expression will be evaluated as a
5705 multiplier value for the coefficient.
5707 If this is option is set, the @option{sigma} option will be ignored.
5709 The absolute value of the coefficient can be accessed through the @var{c}
5713 Set the @var{blocksize} using the number of bits. @code{1<<@var{n}} defines the
5714 @var{blocksize}, which is the width and height of the processed blocks.
5716 The default value is @var{3} (8x8) and can be raised to @var{4} for a
5717 @var{blocksize} of 16x16. Note that changing this setting has huge consequences
5718 on the speed processing. Also, a larger block size does not necessarily means a
5722 @subsection Examples
5724 Apply a denoise with a @option{sigma} of @code{4.5}:
5729 The same operation can be achieved using the expression system:
5731 dctdnoiz=e='gte(c, 4.5*3)'
5734 Violent denoise using a block size of @code{16x16}:
5741 Remove banding artifacts from input video.
5742 It works by replacing banded pixels with average value of referenced pixels.
5744 The filter accepts the following options:
5751 Set banding detection threshold for each plane. Default is 0.02.
5752 Valid range is 0.00003 to 0.5.
5753 If difference between current pixel and reference pixel is less than threshold,
5754 it will be considered as banded.
5757 Banding detection range in pixels. Default is 16. If positive, random number
5758 in range 0 to set value will be used. If negative, exact absolute value
5760 The range defines square of four pixels around current pixel.
5763 Set direction in radians from which four pixel will be compared. If positive,
5764 random direction from 0 to set direction will be picked. If negative, exact of
5765 absolute value will be picked. For example direction 0, -PI or -2*PI radians
5766 will pick only pixels on same row and -PI/2 will pick only pixels on same
5770 If enabled, current pixel is compared with average value of all four
5771 surrounding pixels. The default is enabled. If disabled current pixel is
5772 compared with all four surrounding pixels. The pixel is considered banded
5773 if only all four differences with surrounding pixels are less than threshold.
5779 Drop duplicated frames at regular intervals.
5781 The filter accepts the following options:
5785 Set the number of frames from which one will be dropped. Setting this to
5786 @var{N} means one frame in every batch of @var{N} frames will be dropped.
5787 Default is @code{5}.
5790 Set the threshold for duplicate detection. If the difference metric for a frame
5791 is less than or equal to this value, then it is declared as duplicate. Default
5795 Set scene change threshold. Default is @code{15}.
5799 Set the size of the x and y-axis blocks used during metric calculations.
5800 Larger blocks give better noise suppression, but also give worse detection of
5801 small movements. Must be a power of two. Default is @code{32}.
5804 Mark main input as a pre-processed input and activate clean source input
5805 stream. This allows the input to be pre-processed with various filters to help
5806 the metrics calculation while keeping the frame selection lossless. When set to
5807 @code{1}, the first stream is for the pre-processed input, and the second
5808 stream is the clean source from where the kept frames are chosen. Default is
5812 Set whether or not chroma is considered in the metric calculations. Default is
5818 Apply deflate effect to the video.
5820 This filter replaces the pixel by the local(3x3) average by taking into account
5821 only values lower than the pixel.
5823 It accepts the following options:
5830 Limit the maximum change for each plane, default is 65535.
5831 If 0, plane will remain unchanged.
5836 Remove judder produced by partially interlaced telecined content.
5838 Judder can be introduced, for instance, by @ref{pullup} filter. If the original
5839 source was partially telecined content then the output of @code{pullup,dejudder}
5840 will have a variable frame rate. May change the recorded frame rate of the
5841 container. Aside from that change, this filter will not affect constant frame
5844 The option available in this filter is:
5848 Specify the length of the window over which the judder repeats.
5850 Accepts any integer greater than 1. Useful values are:
5854 If the original was telecined from 24 to 30 fps (Film to NTSC).
5857 If the original was telecined from 25 to 30 fps (PAL to NTSC).
5860 If a mixture of the two.
5863 The default is @samp{4}.
5868 Suppress a TV station logo by a simple interpolation of the surrounding
5869 pixels. Just set a rectangle covering the logo and watch it disappear
5870 (and sometimes something even uglier appear - your mileage may vary).
5872 It accepts the following parameters:
5877 Specify the top left corner coordinates of the logo. They must be
5882 Specify the width and height of the logo to clear. They must be
5886 Specify the thickness of the fuzzy edge of the rectangle (added to
5887 @var{w} and @var{h}). The default value is 1. This option is
5888 deprecated, setting higher values should no longer be necessary and
5892 When set to 1, a green rectangle is drawn on the screen to simplify
5893 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
5894 The default value is 0.
5896 The rectangle is drawn on the outermost pixels which will be (partly)
5897 replaced with interpolated values. The values of the next pixels
5898 immediately outside this rectangle in each direction will be used to
5899 compute the interpolated pixel values inside the rectangle.
5903 @subsection Examples
5907 Set a rectangle covering the area with top left corner coordinates 0,0
5908 and size 100x77, and a band of size 10:
5910 delogo=x=0:y=0:w=100:h=77:band=10
5917 Attempt to fix small changes in horizontal and/or vertical shift. This
5918 filter helps remove camera shake from hand-holding a camera, bumping a
5919 tripod, moving on a vehicle, etc.
5921 The filter accepts the following options:
5929 Specify a rectangular area where to limit the search for motion
5931 If desired the search for motion vectors can be limited to a
5932 rectangular area of the frame defined by its top left corner, width
5933 and height. These parameters have the same meaning as the drawbox
5934 filter which can be used to visualise the position of the bounding
5937 This is useful when simultaneous movement of subjects within the frame
5938 might be confused for camera motion by the motion vector search.
5940 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
5941 then the full frame is used. This allows later options to be set
5942 without specifying the bounding box for the motion vector search.
5944 Default - search the whole frame.
5948 Specify the maximum extent of movement in x and y directions in the
5949 range 0-64 pixels. Default 16.
5952 Specify how to generate pixels to fill blanks at the edge of the
5953 frame. Available values are:
5956 Fill zeroes at blank locations
5958 Original image at blank locations
5960 Extruded edge value at blank locations
5962 Mirrored edge at blank locations
5964 Default value is @samp{mirror}.
5967 Specify the blocksize to use for motion search. Range 4-128 pixels,
5971 Specify the contrast threshold for blocks. Only blocks with more than
5972 the specified contrast (difference between darkest and lightest
5973 pixels) will be considered. Range 1-255, default 125.
5976 Specify the search strategy. Available values are:
5979 Set exhaustive search
5981 Set less exhaustive search.
5983 Default value is @samp{exhaustive}.
5986 If set then a detailed log of the motion search is written to the
5990 If set to 1, specify using OpenCL capabilities, only available if
5991 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
5997 Apply an exact inverse of the telecine operation. It requires a predefined
5998 pattern specified using the pattern option which must be the same as that passed
5999 to the telecine filter.
6001 This filter accepts the following options:
6010 The default value is @code{top}.
6014 A string of numbers representing the pulldown pattern you wish to apply.
6015 The default value is @code{23}.
6018 A number representing position of the first frame with respect to the telecine
6019 pattern. This is to be used if the stream is cut. The default value is @code{0}.
6024 Apply dilation effect to the video.
6026 This filter replaces the pixel by the local(3x3) maximum.
6028 It accepts the following options:
6035 Limit the maximum change for each plane, default is 65535.
6036 If 0, plane will remain unchanged.
6039 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
6042 Flags to local 3x3 coordinates maps like this:
6051 Displace pixels as indicated by second and third input stream.
6053 It takes three input streams and outputs one stream, the first input is the
6054 source, and second and third input are displacement maps.
6056 The second input specifies how much to displace pixels along the
6057 x-axis, while the third input specifies how much to displace pixels
6059 If one of displacement map streams terminates, last frame from that
6060 displacement map will be used.
6062 Note that once generated, displacements maps can be reused over and over again.
6064 A description of the accepted options follows.
6068 Set displace behavior for pixels that are out of range.
6070 Available values are:
6073 Missing pixels are replaced by black pixels.
6076 Adjacent pixels will spread out to replace missing pixels.
6079 Out of range pixels are wrapped so they point to pixels of other side.
6081 Default is @samp{smear}.
6085 @subsection Examples
6089 Add ripple effect to rgb input of video size hd720:
6091 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
6095 Add wave effect to rgb input of video size hd720:
6097 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
6103 Draw a colored box on the input image.
6105 It accepts the following parameters:
6110 The expressions which specify the top left corner coordinates of the box. It defaults to 0.
6114 The expressions which specify the width and height of the box; if 0 they are interpreted as
6115 the input width and height. It defaults to 0.
6118 Specify the color of the box to write. For the general syntax of this option,
6119 check the "Color" section in the ffmpeg-utils manual. If the special
6120 value @code{invert} is used, the box edge color is the same as the
6121 video with inverted luma.
6124 The expression which sets the thickness of the box edge. Default value is @code{3}.
6126 See below for the list of accepted constants.
6129 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6130 following constants:
6134 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6138 horizontal and vertical chroma subsample values. For example for the
6139 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6143 The input width and height.
6146 The input sample aspect ratio.
6150 The x and y offset coordinates where the box is drawn.
6154 The width and height of the drawn box.
6157 The thickness of the drawn box.
6159 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6160 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6164 @subsection Examples
6168 Draw a black box around the edge of the input image:
6174 Draw a box with color red and an opacity of 50%:
6176 drawbox=10:20:200:60:red@@0.5
6179 The previous example can be specified as:
6181 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
6185 Fill the box with pink color:
6187 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
6191 Draw a 2-pixel red 2.40:1 mask:
6193 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
6197 @section drawgraph, adrawgraph
6199 Draw a graph using input video or audio metadata.
6201 It accepts the following parameters:
6205 Set 1st frame metadata key from which metadata values will be used to draw a graph.
6208 Set 1st foreground color expression.
6211 Set 2nd frame metadata key from which metadata values will be used to draw a graph.
6214 Set 2nd foreground color expression.
6217 Set 3rd frame metadata key from which metadata values will be used to draw a graph.
6220 Set 3rd foreground color expression.
6223 Set 4th frame metadata key from which metadata values will be used to draw a graph.
6226 Set 4th foreground color expression.
6229 Set minimal value of metadata value.
6232 Set maximal value of metadata value.
6235 Set graph background color. Default is white.
6240 Available values for mode is:
6247 Default is @code{line}.
6252 Available values for slide is:
6255 Draw new frame when right border is reached.
6258 Replace old columns with new ones.
6261 Scroll from right to left.
6264 Scroll from left to right.
6267 Default is @code{frame}.
6270 Set size of graph video. For the syntax of this option, check the
6271 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
6272 The default value is @code{900x256}.
6274 The foreground color expressions can use the following variables:
6277 Minimal value of metadata value.
6280 Maximal value of metadata value.
6283 Current metadata key value.
6286 The color is defined as 0xAABBGGRR.
6289 Example using metadata from @ref{signalstats} filter:
6291 signalstats,drawgraph=lavfi.signalstats.YAVG:min=0:max=255
6294 Example using metadata from @ref{ebur128} filter:
6296 ebur128=metadata=1,adrawgraph=lavfi.r128.M:min=-120:max=5
6301 Draw a grid on the input image.
6303 It accepts the following parameters:
6308 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
6312 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
6313 input width and height, respectively, minus @code{thickness}, so image gets
6314 framed. Default to 0.
6317 Specify the color of the grid. For the general syntax of this option,
6318 check the "Color" section in the ffmpeg-utils manual. If the special
6319 value @code{invert} is used, the grid color is the same as the
6320 video with inverted luma.
6323 The expression which sets the thickness of the grid line. Default value is @code{1}.
6325 See below for the list of accepted constants.
6328 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
6329 following constants:
6333 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
6337 horizontal and vertical chroma subsample values. For example for the
6338 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6342 The input grid cell width and height.
6345 The input sample aspect ratio.
6349 The x and y coordinates of some point of grid intersection (meant to configure offset).
6353 The width and height of the drawn cell.
6356 The thickness of the drawn cell.
6358 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
6359 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
6363 @subsection Examples
6367 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
6369 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
6373 Draw a white 3x3 grid with an opacity of 50%:
6375 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
6382 Draw a text string or text from a specified file on top of a video, using the
6383 libfreetype library.
6385 To enable compilation of this filter, you need to configure FFmpeg with
6386 @code{--enable-libfreetype}.
6387 To enable default font fallback and the @var{font} option you need to
6388 configure FFmpeg with @code{--enable-libfontconfig}.
6389 To enable the @var{text_shaping} option, you need to configure FFmpeg with
6390 @code{--enable-libfribidi}.
6394 It accepts the following parameters:
6399 Used to draw a box around text using the background color.
6400 The value must be either 1 (enable) or 0 (disable).
6401 The default value of @var{box} is 0.
6404 Set the width of the border to be drawn around the box using @var{boxcolor}.
6405 The default value of @var{boxborderw} is 0.
6408 The color to be used for drawing box around text. For the syntax of this
6409 option, check the "Color" section in the ffmpeg-utils manual.
6411 The default value of @var{boxcolor} is "white".
6414 Set the width of the border to be drawn around the text using @var{bordercolor}.
6415 The default value of @var{borderw} is 0.
6418 Set the color to be used for drawing border around text. For the syntax of this
6419 option, check the "Color" section in the ffmpeg-utils manual.
6421 The default value of @var{bordercolor} is "black".
6424 Select how the @var{text} is expanded. Can be either @code{none},
6425 @code{strftime} (deprecated) or
6426 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
6430 If true, check and fix text coords to avoid clipping.
6433 The color to be used for drawing fonts. For the syntax of this option, check
6434 the "Color" section in the ffmpeg-utils manual.
6436 The default value of @var{fontcolor} is "black".
6438 @item fontcolor_expr
6439 String which is expanded the same way as @var{text} to obtain dynamic
6440 @var{fontcolor} value. By default this option has empty value and is not
6441 processed. When this option is set, it overrides @var{fontcolor} option.
6444 The font family to be used for drawing text. By default Sans.
6447 The font file to be used for drawing text. The path must be included.
6448 This parameter is mandatory if the fontconfig support is disabled.
6451 This option does not exist, please see the timeline system
6454 Draw the text applying alpha blending. The value can
6455 be either a number between 0.0 and 1.0
6456 The expression accepts the same variables @var{x, y} do.
6457 The default value is 1.
6458 Please see fontcolor_expr
6461 The font size to be used for drawing text.
6462 The default value of @var{fontsize} is 16.
6465 If set to 1, attempt to shape the text (for example, reverse the order of
6466 right-to-left text and join Arabic characters) before drawing it.
6467 Otherwise, just draw the text exactly as given.
6468 By default 1 (if supported).
6471 The flags to be used for loading the fonts.
6473 The flags map the corresponding flags supported by libfreetype, and are
6474 a combination of the following values:
6481 @item vertical_layout
6482 @item force_autohint
6485 @item ignore_global_advance_width
6487 @item ignore_transform
6493 Default value is "default".
6495 For more information consult the documentation for the FT_LOAD_*
6499 The color to be used for drawing a shadow behind the drawn text. For the
6500 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
6502 The default value of @var{shadowcolor} is "black".
6506 The x and y offsets for the text shadow position with respect to the
6507 position of the text. They can be either positive or negative
6508 values. The default value for both is "0".
6511 The starting frame number for the n/frame_num variable. The default value
6515 The size in number of spaces to use for rendering the tab.
6519 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
6520 format. It can be used with or without text parameter. @var{timecode_rate}
6521 option must be specified.
6523 @item timecode_rate, rate, r
6524 Set the timecode frame rate (timecode only).
6527 The text string to be drawn. The text must be a sequence of UTF-8
6529 This parameter is mandatory if no file is specified with the parameter
6533 A text file containing text to be drawn. The text must be a sequence
6534 of UTF-8 encoded characters.
6536 This parameter is mandatory if no text string is specified with the
6537 parameter @var{text}.
6539 If both @var{text} and @var{textfile} are specified, an error is thrown.
6542 If set to 1, the @var{textfile} will be reloaded before each frame.
6543 Be sure to update it atomically, or it may be read partially, or even fail.
6547 The expressions which specify the offsets where text will be drawn
6548 within the video frame. They are relative to the top/left border of the
6551 The default value of @var{x} and @var{y} is "0".
6553 See below for the list of accepted constants and functions.
6556 The parameters for @var{x} and @var{y} are expressions containing the
6557 following constants and functions:
6561 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
6565 horizontal and vertical chroma subsample values. For example for the
6566 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6569 the height of each text line
6577 @item max_glyph_a, ascent
6578 the maximum distance from the baseline to the highest/upper grid
6579 coordinate used to place a glyph outline point, for all the rendered
6581 It is a positive value, due to the grid's orientation with the Y axis
6584 @item max_glyph_d, descent
6585 the maximum distance from the baseline to the lowest grid coordinate
6586 used to place a glyph outline point, for all the rendered glyphs.
6587 This is a negative value, due to the grid's orientation, with the Y axis
6591 maximum glyph height, that is the maximum height for all the glyphs
6592 contained in the rendered text, it is equivalent to @var{ascent} -
6596 maximum glyph width, that is the maximum width for all the glyphs
6597 contained in the rendered text
6600 the number of input frame, starting from 0
6602 @item rand(min, max)
6603 return a random number included between @var{min} and @var{max}
6606 The input sample aspect ratio.
6609 timestamp expressed in seconds, NAN if the input timestamp is unknown
6612 the height of the rendered text
6615 the width of the rendered text
6619 the x and y offset coordinates where the text is drawn.
6621 These parameters allow the @var{x} and @var{y} expressions to refer
6622 each other, so you can for example specify @code{y=x/dar}.
6625 @anchor{drawtext_expansion}
6626 @subsection Text expansion
6628 If @option{expansion} is set to @code{strftime},
6629 the filter recognizes strftime() sequences in the provided text and
6630 expands them accordingly. Check the documentation of strftime(). This
6631 feature is deprecated.
6633 If @option{expansion} is set to @code{none}, the text is printed verbatim.
6635 If @option{expansion} is set to @code{normal} (which is the default),
6636 the following expansion mechanism is used.
6638 The backslash character @samp{\}, followed by any character, always expands to
6639 the second character.
6641 Sequence of the form @code{%@{...@}} are expanded. The text between the
6642 braces is a function name, possibly followed by arguments separated by ':'.
6643 If the arguments contain special characters or delimiters (':' or '@}'),
6644 they should be escaped.
6646 Note that they probably must also be escaped as the value for the
6647 @option{text} option in the filter argument string and as the filter
6648 argument in the filtergraph description, and possibly also for the shell,
6649 that makes up to four levels of escaping; using a text file avoids these
6652 The following functions are available:
6657 The expression evaluation result.
6659 It must take one argument specifying the expression to be evaluated,
6660 which accepts the same constants and functions as the @var{x} and
6661 @var{y} values. Note that not all constants should be used, for
6662 example the text size is not known when evaluating the expression, so
6663 the constants @var{text_w} and @var{text_h} will have an undefined
6666 @item expr_int_format, eif
6667 Evaluate the expression's value and output as formatted integer.
6669 The first argument is the expression to be evaluated, just as for the @var{expr} function.
6670 The second argument specifies the output format. Allowed values are @samp{x},
6671 @samp{X}, @samp{d} and @samp{u}. They are treated exactly as in the
6672 @code{printf} function.
6673 The third parameter is optional and sets the number of positions taken by the output.
6674 It can be used to add padding with zeros from the left.
6677 The time at which the filter is running, expressed in UTC.
6678 It can accept an argument: a strftime() format string.
6681 The time at which the filter is running, expressed in the local time zone.
6682 It can accept an argument: a strftime() format string.
6685 Frame metadata. Takes one or two arguments.
6687 The first argument is mandatory and specifies the metadata key.
6689 The second argument is optional and specifies a default value, used when the
6690 metadata key is not found or empty.
6693 The frame number, starting from 0.
6696 A 1 character description of the current picture type.
6699 The timestamp of the current frame.
6700 It can take up to three arguments.
6702 The first argument is the format of the timestamp; it defaults to @code{flt}
6703 for seconds as a decimal number with microsecond accuracy; @code{hms} stands
6704 for a formatted @var{[-]HH:MM:SS.mmm} timestamp with millisecond accuracy.
6705 @code{gmtime} stands for the timestamp of the frame formatted as UTC time;
6706 @code{localtime} stands for the timestamp of the frame formatted as
6707 local time zone time.
6709 The second argument is an offset added to the timestamp.
6711 If the format is set to @code{localtime} or @code{gmtime},
6712 a third argument may be supplied: a strftime() format string.
6713 By default, @var{YYYY-MM-DD HH:MM:SS} format will be used.
6716 @subsection Examples
6720 Draw "Test Text" with font FreeSerif, using the default values for the
6721 optional parameters.
6724 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
6728 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
6729 and y=50 (counting from the top-left corner of the screen), text is
6730 yellow with a red box around it. Both the text and the box have an
6734 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
6735 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
6738 Note that the double quotes are not necessary if spaces are not used
6739 within the parameter list.
6742 Show the text at the center of the video frame:
6744 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h)/2"
6748 Show the text at a random position, switching to a new position every 30 seconds:
6750 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)"
6754 Show a text line sliding from right to left in the last row of the video
6755 frame. The file @file{LONG_LINE} is assumed to contain a single line
6758 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
6762 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
6764 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
6768 Draw a single green letter "g", at the center of the input video.
6769 The glyph baseline is placed at half screen height.
6771 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
6775 Show text for 1 second every 3 seconds:
6777 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
6781 Use fontconfig to set the font. Note that the colons need to be escaped.
6783 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
6787 Print the date of a real-time encoding (see strftime(3)):
6789 drawtext='fontfile=FreeSans.ttf:text=%@{localtime\:%a %b %d %Y@}'
6793 Show text fading in and out (appearing/disappearing):
6796 DS=1.0 # display start
6797 DE=10.0 # display end
6798 FID=1.5 # fade in duration
6799 FOD=5 # fade out duration
6800 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 @}"
6805 For more information about libfreetype, check:
6806 @url{http://www.freetype.org/}.
6808 For more information about fontconfig, check:
6809 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
6811 For more information about libfribidi, check:
6812 @url{http://fribidi.org/}.
6816 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
6818 The filter accepts the following options:
6823 Set low and high threshold values used by the Canny thresholding
6826 The high threshold selects the "strong" edge pixels, which are then
6827 connected through 8-connectivity with the "weak" edge pixels selected
6828 by the low threshold.
6830 @var{low} and @var{high} threshold values must be chosen in the range
6831 [0,1], and @var{low} should be lesser or equal to @var{high}.
6833 Default value for @var{low} is @code{20/255}, and default value for @var{high}
6837 Define the drawing mode.
6841 Draw white/gray wires on black background.
6844 Mix the colors to create a paint/cartoon effect.
6847 Default value is @var{wires}.
6850 @subsection Examples
6854 Standard edge detection with custom values for the hysteresis thresholding:
6856 edgedetect=low=0.1:high=0.4
6860 Painting effect without thresholding:
6862 edgedetect=mode=colormix:high=0
6867 Set brightness, contrast, saturation and approximate gamma adjustment.
6869 The filter accepts the following options:
6873 Set the contrast expression. The value must be a float value in range
6874 @code{-2.0} to @code{2.0}. The default value is "1".
6877 Set the brightness expression. The value must be a float value in
6878 range @code{-1.0} to @code{1.0}. The default value is "0".
6881 Set the saturation expression. The value must be a float in
6882 range @code{0.0} to @code{3.0}. The default value is "1".
6885 Set the gamma expression. The value must be a float in range
6886 @code{0.1} to @code{10.0}. The default value is "1".
6889 Set the gamma expression for red. The value must be a float in
6890 range @code{0.1} to @code{10.0}. The default value is "1".
6893 Set the gamma expression for green. The value must be a float in range
6894 @code{0.1} to @code{10.0}. The default value is "1".
6897 Set the gamma expression for blue. The value must be a float in range
6898 @code{0.1} to @code{10.0}. The default value is "1".
6901 Set the gamma weight expression. It can be used to reduce the effect
6902 of a high gamma value on bright image areas, e.g. keep them from
6903 getting overamplified and just plain white. The value must be a float
6904 in range @code{0.0} to @code{1.0}. A value of @code{0.0} turns the
6905 gamma correction all the way down while @code{1.0} leaves it at its
6906 full strength. Default is "1".
6909 Set when the expressions for brightness, contrast, saturation and
6910 gamma expressions are evaluated.
6912 It accepts the following values:
6915 only evaluate expressions once during the filter initialization or
6916 when a command is processed
6919 evaluate expressions for each incoming frame
6922 Default value is @samp{init}.
6925 The expressions accept the following parameters:
6928 frame count of the input frame starting from 0
6931 byte position of the corresponding packet in the input file, NAN if
6935 frame rate of the input video, NAN if the input frame rate is unknown
6938 timestamp expressed in seconds, NAN if the input timestamp is unknown
6941 @subsection Commands
6942 The filter supports the following commands:
6946 Set the contrast expression.
6949 Set the brightness expression.
6952 Set the saturation expression.
6955 Set the gamma expression.
6958 Set the gamma_r expression.
6961 Set gamma_g expression.
6964 Set gamma_b expression.
6967 Set gamma_weight expression.
6969 The command accepts the same syntax of the corresponding option.
6971 If the specified expression is not valid, it is kept at its current
6978 Apply erosion effect to the video.
6980 This filter replaces the pixel by the local(3x3) minimum.
6982 It accepts the following options:
6989 Limit the maximum change for each plane, default is 65535.
6990 If 0, plane will remain unchanged.
6993 Flag which specifies the pixel to refer to. Default is 255 i.e. all eight
6996 Flags to local 3x3 coordinates maps like this:
7003 @section extractplanes
7005 Extract color channel components from input video stream into
7006 separate grayscale video streams.
7008 The filter accepts the following option:
7012 Set plane(s) to extract.
7014 Available values for planes are:
7025 Choosing planes not available in the input will result in an error.
7026 That means you cannot select @code{r}, @code{g}, @code{b} planes
7027 with @code{y}, @code{u}, @code{v} planes at same time.
7030 @subsection Examples
7034 Extract luma, u and v color channel component from input video frame
7035 into 3 grayscale outputs:
7037 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
7043 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
7045 For each input image, the filter will compute the optimal mapping from
7046 the input to the output given the codebook length, that is the number
7047 of distinct output colors.
7049 This filter accepts the following options.
7052 @item codebook_length, l
7053 Set codebook length. The value must be a positive integer, and
7054 represents the number of distinct output colors. Default value is 256.
7057 Set the maximum number of iterations to apply for computing the optimal
7058 mapping. The higher the value the better the result and the higher the
7059 computation time. Default value is 1.
7062 Set a random seed, must be an integer included between 0 and
7063 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
7064 will try to use a good random seed on a best effort basis.
7067 Set pal8 output pixel format. This option does not work with codebook
7068 length greater than 256.
7073 Apply a fade-in/out effect to the input video.
7075 It accepts the following parameters:
7079 The effect type can be either "in" for a fade-in, or "out" for a fade-out
7081 Default is @code{in}.
7083 @item start_frame, s
7084 Specify the number of the frame to start applying the fade
7085 effect at. Default is 0.
7088 The number of frames that the fade effect lasts. At the end of the
7089 fade-in effect, the output video will have the same intensity as the input video.
7090 At the end of the fade-out transition, the output video will be filled with the
7091 selected @option{color}.
7095 If set to 1, fade only alpha channel, if one exists on the input.
7098 @item start_time, st
7099 Specify the timestamp (in seconds) of the frame to start to apply the fade
7100 effect. If both start_frame and start_time are specified, the fade will start at
7101 whichever comes last. Default is 0.
7104 The number of seconds for which the fade effect has to last. At the end of the
7105 fade-in effect the output video will have the same intensity as the input video,
7106 at the end of the fade-out transition the output video will be filled with the
7107 selected @option{color}.
7108 If both duration and nb_frames are specified, duration is used. Default is 0
7109 (nb_frames is used by default).
7112 Specify the color of the fade. Default is "black".
7115 @subsection Examples
7119 Fade in the first 30 frames of video:
7124 The command above is equivalent to:
7130 Fade out the last 45 frames of a 200-frame video:
7133 fade=type=out:start_frame=155:nb_frames=45
7137 Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
7139 fade=in:0:25, fade=out:975:25
7143 Make the first 5 frames yellow, then fade in from frame 5-24:
7145 fade=in:5:20:color=yellow
7149 Fade in alpha over first 25 frames of video:
7151 fade=in:0:25:alpha=1
7155 Make the first 5.5 seconds black, then fade in for 0.5 seconds:
7157 fade=t=in:st=5.5:d=0.5
7163 Apply arbitrary expressions to samples in frequency domain
7167 Adjust the dc value (gain) of the luma plane of the image. The filter
7168 accepts an integer value in range @code{0} to @code{1000}. The default
7169 value is set to @code{0}.
7172 Adjust the dc value (gain) of the 1st chroma plane of the image. The
7173 filter accepts an integer value in range @code{0} to @code{1000}. The
7174 default value is set to @code{0}.
7177 Adjust the dc value (gain) of the 2nd chroma plane of the image. The
7178 filter accepts an integer value in range @code{0} to @code{1000}. The
7179 default value is set to @code{0}.
7182 Set the frequency domain weight expression for the luma plane.
7185 Set the frequency domain weight expression for the 1st chroma plane.
7188 Set the frequency domain weight expression for the 2nd chroma plane.
7190 The filter accepts the following variables:
7193 The coordinates of the current sample.
7197 The width and height of the image.
7200 @subsection Examples
7206 fftfilt=dc_Y=128:weight_Y='squish(1-(Y+X)/100)'
7212 fftfilt=dc_Y=0:weight_Y='squish((Y+X)/100-1)'
7218 fftfilt=dc_Y=0:weight_Y='1+squish(1-(Y+X)/100)'
7224 fftfilt=dc_Y=0:weight_Y='exp(-4 * ((Y+X)/(W+H)))'
7231 Extract a single field from an interlaced image using stride
7232 arithmetic to avoid wasting CPU time. The output frames are marked as
7235 The filter accepts the following options:
7239 Specify whether to extract the top (if the value is @code{0} or
7240 @code{top}) or the bottom field (if the value is @code{1} or
7246 Create new frames by copying the top and bottom fields from surrounding frames
7247 supplied as numbers by the hint file.
7251 Set file containing hints: absolute/relative frame numbers.
7253 There must be one line for each frame in a clip. Each line must contain two
7254 numbers separated by the comma, optionally followed by @code{-} or @code{+}.
7255 Numbers supplied on each line of file can not be out of [N-1,N+1] where N
7256 is current frame number for @code{absolute} mode or out of [-1, 1] range
7257 for @code{relative} mode. First number tells from which frame to pick up top
7258 field and second number tells from which frame to pick up bottom field.
7260 If optionally followed by @code{+} output frame will be marked as interlaced,
7261 else if followed by @code{-} output frame will be marked as progressive, else
7262 it will be marked same as input frame.
7263 If line starts with @code{#} or @code{;} that line is skipped.
7266 Can be item @code{absolute} or @code{relative}. Default is @code{absolute}.
7269 Example of first several lines of @code{hint} file for @code{relative} mode:
7272 1,0 - # second frame, use third's frame top field and second's frame bottom field
7273 1,0 - # third frame, use fourth's frame top field and third's frame bottom field
7290 Field matching filter for inverse telecine. It is meant to reconstruct the
7291 progressive frames from a telecined stream. The filter does not drop duplicated
7292 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
7293 followed by a decimation filter such as @ref{decimate} in the filtergraph.
7295 The separation of the field matching and the decimation is notably motivated by
7296 the possibility of inserting a de-interlacing filter fallback between the two.
7297 If the source has mixed telecined and real interlaced content,
7298 @code{fieldmatch} will not be able to match fields for the interlaced parts.
7299 But these remaining combed frames will be marked as interlaced, and thus can be
7300 de-interlaced by a later filter such as @ref{yadif} before decimation.
7302 In addition to the various configuration options, @code{fieldmatch} can take an
7303 optional second stream, activated through the @option{ppsrc} option. If
7304 enabled, the frames reconstruction will be based on the fields and frames from
7305 this second stream. This allows the first input to be pre-processed in order to
7306 help the various algorithms of the filter, while keeping the output lossless
7307 (assuming the fields are matched properly). Typically, a field-aware denoiser,
7308 or brightness/contrast adjustments can help.
7310 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
7311 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
7312 which @code{fieldmatch} is based on. While the semantic and usage are very
7313 close, some behaviour and options names can differ.
7315 The @ref{decimate} filter currently only works for constant frame rate input.
7316 If your input has mixed telecined (30fps) and progressive content with a lower
7317 framerate like 24fps use the following filterchain to produce the necessary cfr
7318 stream: @code{dejudder,fps=30000/1001,fieldmatch,decimate}.
7320 The filter accepts the following options:
7324 Specify the assumed field order of the input stream. Available values are:
7328 Auto detect parity (use FFmpeg's internal parity value).
7330 Assume bottom field first.
7332 Assume top field first.
7335 Note that it is sometimes recommended not to trust the parity announced by the
7338 Default value is @var{auto}.
7341 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
7342 sense that it won't risk creating jerkiness due to duplicate frames when
7343 possible, but if there are bad edits or blended fields it will end up
7344 outputting combed frames when a good match might actually exist. On the other
7345 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
7346 but will almost always find a good frame if there is one. The other values are
7347 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
7348 jerkiness and creating duplicate frames versus finding good matches in sections
7349 with bad edits, orphaned fields, blended fields, etc.
7351 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
7353 Available values are:
7357 2-way matching (p/c)
7359 2-way matching, and trying 3rd match if still combed (p/c + n)
7361 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
7363 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
7364 still combed (p/c + n + u/b)
7366 3-way matching (p/c/n)
7368 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
7369 detected as combed (p/c/n + u/b)
7372 The parenthesis at the end indicate the matches that would be used for that
7373 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
7376 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
7379 Default value is @var{pc_n}.
7382 Mark the main input stream as a pre-processed input, and enable the secondary
7383 input stream as the clean source to pick the fields from. See the filter
7384 introduction for more details. It is similar to the @option{clip2} feature from
7387 Default value is @code{0} (disabled).
7390 Set the field to match from. It is recommended to set this to the same value as
7391 @option{order} unless you experience matching failures with that setting. In
7392 certain circumstances changing the field that is used to match from can have a
7393 large impact on matching performance. Available values are:
7397 Automatic (same value as @option{order}).
7399 Match from the bottom field.
7401 Match from the top field.
7404 Default value is @var{auto}.
7407 Set whether or not chroma is included during the match comparisons. In most
7408 cases it is recommended to leave this enabled. You should set this to @code{0}
7409 only if your clip has bad chroma problems such as heavy rainbowing or other
7410 artifacts. Setting this to @code{0} could also be used to speed things up at
7411 the cost of some accuracy.
7413 Default value is @code{1}.
7417 These define an exclusion band which excludes the lines between @option{y0} and
7418 @option{y1} from being included in the field matching decision. An exclusion
7419 band can be used to ignore subtitles, a logo, or other things that may
7420 interfere with the matching. @option{y0} sets the starting scan line and
7421 @option{y1} sets the ending line; all lines in between @option{y0} and
7422 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
7423 @option{y0} and @option{y1} to the same value will disable the feature.
7424 @option{y0} and @option{y1} defaults to @code{0}.
7427 Set the scene change detection threshold as a percentage of maximum change on
7428 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
7429 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
7430 @option{scthresh} is @code{[0.0, 100.0]}.
7432 Default value is @code{12.0}.
7435 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
7436 account the combed scores of matches when deciding what match to use as the
7437 final match. Available values are:
7441 No final matching based on combed scores.
7443 Combed scores are only used when a scene change is detected.
7445 Use combed scores all the time.
7448 Default is @var{sc}.
7451 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
7452 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
7453 Available values are:
7457 No forced calculation.
7459 Force p/c/n calculations.
7461 Force p/c/n/u/b calculations.
7464 Default value is @var{none}.
7467 This is the area combing threshold used for combed frame detection. This
7468 essentially controls how "strong" or "visible" combing must be to be detected.
7469 Larger values mean combing must be more visible and smaller values mean combing
7470 can be less visible or strong and still be detected. Valid settings are from
7471 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
7472 be detected as combed). This is basically a pixel difference value. A good
7473 range is @code{[8, 12]}.
7475 Default value is @code{9}.
7478 Sets whether or not chroma is considered in the combed frame decision. Only
7479 disable this if your source has chroma problems (rainbowing, etc.) that are
7480 causing problems for the combed frame detection with chroma enabled. Actually,
7481 using @option{chroma}=@var{0} is usually more reliable, except for the case
7482 where there is chroma only combing in the source.
7484 Default value is @code{0}.
7488 Respectively set the x-axis and y-axis size of the window used during combed
7489 frame detection. This has to do with the size of the area in which
7490 @option{combpel} pixels are required to be detected as combed for a frame to be
7491 declared combed. See the @option{combpel} parameter description for more info.
7492 Possible values are any number that is a power of 2 starting at 4 and going up
7495 Default value is @code{16}.
7498 The number of combed pixels inside any of the @option{blocky} by
7499 @option{blockx} size blocks on the frame for the frame to be detected as
7500 combed. While @option{cthresh} controls how "visible" the combing must be, this
7501 setting controls "how much" combing there must be in any localized area (a
7502 window defined by the @option{blockx} and @option{blocky} settings) on the
7503 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
7504 which point no frames will ever be detected as combed). This setting is known
7505 as @option{MI} in TFM/VFM vocabulary.
7507 Default value is @code{80}.
7510 @anchor{p/c/n/u/b meaning}
7511 @subsection p/c/n/u/b meaning
7513 @subsubsection p/c/n
7515 We assume the following telecined stream:
7518 Top fields: 1 2 2 3 4
7519 Bottom fields: 1 2 3 4 4
7522 The numbers correspond to the progressive frame the fields relate to. Here, the
7523 first two frames are progressive, the 3rd and 4th are combed, and so on.
7525 When @code{fieldmatch} is configured to run a matching from bottom
7526 (@option{field}=@var{bottom}) this is how this input stream get transformed:
7531 B 1 2 3 4 4 <-- matching reference
7540 As a result of the field matching, we can see that some frames get duplicated.
7541 To perform a complete inverse telecine, you need to rely on a decimation filter
7542 after this operation. See for instance the @ref{decimate} filter.
7544 The same operation now matching from top fields (@option{field}=@var{top})
7549 T 1 2 2 3 4 <-- matching reference
7559 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
7560 basically, they refer to the frame and field of the opposite parity:
7563 @item @var{p} matches the field of the opposite parity in the previous frame
7564 @item @var{c} matches the field of the opposite parity in the current frame
7565 @item @var{n} matches the field of the opposite parity in the next frame
7570 The @var{u} and @var{b} matching are a bit special in the sense that they match
7571 from the opposite parity flag. In the following examples, we assume that we are
7572 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
7573 'x' is placed above and below each matched fields.
7575 With bottom matching (@option{field}=@var{bottom}):
7580 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7581 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
7589 With top matching (@option{field}=@var{top}):
7594 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
7595 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
7603 @subsection Examples
7605 Simple IVTC of a top field first telecined stream:
7607 fieldmatch=order=tff:combmatch=none, decimate
7610 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
7612 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
7617 Transform the field order of the input video.
7619 It accepts the following parameters:
7624 The output field order. Valid values are @var{tff} for top field first or @var{bff}
7625 for bottom field first.
7628 The default value is @samp{tff}.
7630 The transformation is done by shifting the picture content up or down
7631 by one line, and filling the remaining line with appropriate picture content.
7632 This method is consistent with most broadcast field order converters.
7634 If the input video is not flagged as being interlaced, or it is already
7635 flagged as being of the required output field order, then this filter does
7636 not alter the incoming video.
7638 It is very useful when converting to or from PAL DV material,
7639 which is bottom field first.
7643 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
7646 @section fifo, afifo
7648 Buffer input images and send them when they are requested.
7650 It is mainly useful when auto-inserted by the libavfilter
7653 It does not take parameters.
7657 Find a rectangular object
7659 It accepts the following options:
7663 Filepath of the object image, needs to be in gray8.
7666 Detection threshold, default is 0.5.
7669 Number of mipmaps, default is 3.
7671 @item xmin, ymin, xmax, ymax
7672 Specifies the rectangle in which to search.
7675 @subsection Examples
7679 Generate a representative palette of a given video using @command{ffmpeg}:
7681 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
7687 Cover a rectangular object
7689 It accepts the following options:
7693 Filepath of the optional cover image, needs to be in yuv420.
7698 It accepts the following values:
7701 cover it by the supplied image
7703 cover it by interpolating the surrounding pixels
7706 Default value is @var{blur}.
7709 @subsection Examples
7713 Generate a representative palette of a given video using @command{ffmpeg}:
7715 ffmpeg -i file.ts -vf find_rect=newref.pgm,cover_rect=cover.jpg:mode=cover new.mkv
7722 Convert the input video to one of the specified pixel formats.
7723 Libavfilter will try to pick one that is suitable as input to
7726 It accepts the following parameters:
7730 A '|'-separated list of pixel format names, such as
7731 "pix_fmts=yuv420p|monow|rgb24".
7735 @subsection Examples
7739 Convert the input video to the @var{yuv420p} format
7741 format=pix_fmts=yuv420p
7744 Convert the input video to any of the formats in the list
7746 format=pix_fmts=yuv420p|yuv444p|yuv410p
7753 Convert the video to specified constant frame rate by duplicating or dropping
7754 frames as necessary.
7756 It accepts the following parameters:
7760 The desired output frame rate. The default is @code{25}.
7765 Possible values are:
7768 zero round towards 0
7772 round towards -infinity
7774 round towards +infinity
7778 The default is @code{near}.
7781 Assume the first PTS should be the given value, in seconds. This allows for
7782 padding/trimming at the start of stream. By default, no assumption is made
7783 about the first frame's expected PTS, so no padding or trimming is done.
7784 For example, this could be set to 0 to pad the beginning with duplicates of
7785 the first frame if a video stream starts after the audio stream or to trim any
7786 frames with a negative PTS.
7790 Alternatively, the options can be specified as a flat string:
7791 @var{fps}[:@var{round}].
7793 See also the @ref{setpts} filter.
7795 @subsection Examples
7799 A typical usage in order to set the fps to 25:
7805 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
7807 fps=fps=film:round=near
7813 Pack two different video streams into a stereoscopic video, setting proper
7814 metadata on supported codecs. The two views should have the same size and
7815 framerate and processing will stop when the shorter video ends. Please note
7816 that you may conveniently adjust view properties with the @ref{scale} and
7819 It accepts the following parameters:
7823 The desired packing format. Supported values are:
7828 The views are next to each other (default).
7831 The views are on top of each other.
7834 The views are packed by line.
7837 The views are packed by column.
7840 The views are temporally interleaved.
7849 # Convert left and right views into a frame-sequential video
7850 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
7852 # Convert views into a side-by-side video with the same output resolution as the input
7853 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
7858 Change the frame rate by interpolating new video output frames from the source
7861 This filter is not designed to function correctly with interlaced media. If
7862 you wish to change the frame rate of interlaced media then you are required
7863 to deinterlace before this filter and re-interlace after this filter.
7865 A description of the accepted options follows.
7869 Specify the output frames per second. This option can also be specified
7870 as a value alone. The default is @code{50}.
7873 Specify the start of a range where the output frame will be created as a
7874 linear interpolation of two frames. The range is [@code{0}-@code{255}],
7875 the default is @code{15}.
7878 Specify the end of a range where the output frame will be created as a
7879 linear interpolation of two frames. The range is [@code{0}-@code{255}],
7880 the default is @code{240}.
7883 Specify the level at which a scene change is detected as a value between
7884 0 and 100 to indicate a new scene; a low value reflects a low
7885 probability for the current frame to introduce a new scene, while a higher
7886 value means the current frame is more likely to be one.
7887 The default is @code{7}.
7890 Specify flags influencing the filter process.
7892 Available value for @var{flags} is:
7895 @item scene_change_detect, scd
7896 Enable scene change detection using the value of the option @var{scene}.
7897 This flag is enabled by default.
7903 Select one frame every N-th frame.
7905 This filter accepts the following option:
7908 Select frame after every @code{step} frames.
7909 Allowed values are positive integers higher than 0. Default value is @code{1}.
7915 Apply a frei0r effect to the input video.
7917 To enable the compilation of this filter, you need to install the frei0r
7918 header and configure FFmpeg with @code{--enable-frei0r}.
7920 It accepts the following parameters:
7925 The name of the frei0r effect to load. If the environment variable
7926 @env{FREI0R_PATH} is defined, the frei0r effect is searched for in each of the
7927 directories specified by the colon-separated list in @env{FREIOR_PATH}.
7928 Otherwise, the standard frei0r paths are searched, in this order:
7929 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
7930 @file{/usr/lib/frei0r-1/}.
7933 A '|'-separated list of parameters to pass to the frei0r effect.
7937 A frei0r effect parameter can be a boolean (its value is either
7938 "y" or "n"), a double, a color (specified as
7939 @var{R}/@var{G}/@var{B}, where @var{R}, @var{G}, and @var{B} are floating point
7940 numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
7941 section in the ffmpeg-utils manual), a position (specified as @var{X}/@var{Y}, where
7942 @var{X} and @var{Y} are floating point numbers) and/or a string.
7944 The number and types of parameters depend on the loaded effect. If an
7945 effect parameter is not specified, the default value is set.
7947 @subsection Examples
7951 Apply the distort0r effect, setting the first two double parameters:
7953 frei0r=filter_name=distort0r:filter_params=0.5|0.01
7957 Apply the colordistance effect, taking a color as the first parameter:
7959 frei0r=colordistance:0.2/0.3/0.4
7960 frei0r=colordistance:violet
7961 frei0r=colordistance:0x112233
7965 Apply the perspective effect, specifying the top left and top right image
7968 frei0r=perspective:0.2/0.2|0.8/0.2
7972 For more information, see
7973 @url{http://frei0r.dyne.org}
7977 Apply fast and simple postprocessing. It is a faster version of @ref{spp}.
7979 It splits (I)DCT into horizontal/vertical passes. Unlike the simple post-
7980 processing filter, one of them is performed once per block, not per pixel.
7981 This allows for much higher speed.
7983 The filter accepts the following options:
7987 Set quality. This option defines the number of levels for averaging. It accepts
7988 an integer in the range 4-5. Default value is @code{4}.
7991 Force a constant quantization parameter. It accepts an integer in range 0-63.
7992 If not set, the filter will use the QP from the video stream (if available).
7995 Set filter strength. It accepts an integer in range -15 to 32. Lower values mean
7996 more details but also more artifacts, while higher values make the image smoother
7997 but also blurrier. Default value is @code{0} − PSNR optimal.
8000 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
8001 option may cause flicker since the B-Frames have often larger QP. Default is
8002 @code{0} (not enabled).
8008 The filter accepts the following options:
8012 Set the luminance expression.
8014 Set the chrominance blue expression.
8016 Set the chrominance red expression.
8018 Set the alpha expression.
8020 Set the red expression.
8022 Set the green expression.
8024 Set the blue expression.
8027 The colorspace is selected according to the specified options. If one
8028 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
8029 options is specified, the filter will automatically select a YCbCr
8030 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
8031 @option{blue_expr} options is specified, it will select an RGB
8034 If one of the chrominance expression is not defined, it falls back on the other
8035 one. If no alpha expression is specified it will evaluate to opaque value.
8036 If none of chrominance expressions are specified, they will evaluate
8037 to the luminance expression.
8039 The expressions can use the following variables and functions:
8043 The sequential number of the filtered frame, starting from @code{0}.
8047 The coordinates of the current sample.
8051 The width and height of the image.
8055 Width and height scale depending on the currently filtered plane. It is the
8056 ratio between the corresponding luma plane number of pixels and the current
8057 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
8058 @code{0.5,0.5} for chroma planes.
8061 Time of the current frame, expressed in seconds.
8064 Return the value of the pixel at location (@var{x},@var{y}) of the current
8068 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
8072 Return the value of the pixel at location (@var{x},@var{y}) of the
8073 blue-difference chroma plane. Return 0 if there is no such plane.
8076 Return the value of the pixel at location (@var{x},@var{y}) of the
8077 red-difference chroma plane. Return 0 if there is no such plane.
8082 Return the value of the pixel at location (@var{x},@var{y}) of the
8083 red/green/blue component. Return 0 if there is no such component.
8086 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
8087 plane. Return 0 if there is no such plane.
8090 For functions, if @var{x} and @var{y} are outside the area, the value will be
8091 automatically clipped to the closer edge.
8093 @subsection Examples
8097 Flip the image horizontally:
8103 Generate a bidimensional sine wave, with angle @code{PI/3} and a
8104 wavelength of 100 pixels:
8106 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
8110 Generate a fancy enigmatic moving light:
8112 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
8116 Generate a quick emboss effect:
8118 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
8122 Modify RGB components depending on pixel position:
8124 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
8128 Create a radial gradient that is the same size as the input (also see
8129 the @ref{vignette} filter):
8131 geq=lum=255*gauss((X/W-0.5)*3)*gauss((Y/H-0.5)*3)/gauss(0)/gauss(0),format=gray
8137 Fix the banding artifacts that are sometimes introduced into nearly flat
8138 regions by truncation to 8bit color depth.
8139 Interpolate the gradients that should go where the bands are, and
8142 It is designed for playback only. Do not use it prior to
8143 lossy compression, because compression tends to lose the dither and
8144 bring back the bands.
8146 It accepts the following parameters:
8151 The maximum amount by which the filter will change any one pixel. This is also
8152 the threshold for detecting nearly flat regions. Acceptable values range from
8153 .51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
8157 The neighborhood to fit the gradient to. A larger radius makes for smoother
8158 gradients, but also prevents the filter from modifying the pixels near detailed
8159 regions. Acceptable values are 8-32; the default value is 16. Out-of-range
8160 values will be clipped to the valid range.
8164 Alternatively, the options can be specified as a flat string:
8165 @var{strength}[:@var{radius}]
8167 @subsection Examples
8171 Apply the filter with a @code{3.5} strength and radius of @code{8}:
8177 Specify radius, omitting the strength (which will fall-back to the default
8188 Apply a Hald CLUT to a video stream.
8190 First input is the video stream to process, and second one is the Hald CLUT.
8191 The Hald CLUT input can be a simple picture or a complete video stream.
8193 The filter accepts the following options:
8197 Force termination when the shortest input terminates. Default is @code{0}.
8199 Continue applying the last CLUT after the end of the stream. A value of
8200 @code{0} disable the filter after the last frame of the CLUT is reached.
8201 Default is @code{1}.
8204 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
8205 filters share the same internals).
8207 More information about the Hald CLUT can be found on Eskil Steenberg's website
8208 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
8210 @subsection Workflow examples
8212 @subsubsection Hald CLUT video stream
8214 Generate an identity Hald CLUT stream altered with various effects:
8216 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
8219 Note: make sure you use a lossless codec.
8221 Then use it with @code{haldclut} to apply it on some random stream:
8223 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
8226 The Hald CLUT will be applied to the 10 first seconds (duration of
8227 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
8228 to the remaining frames of the @code{mandelbrot} stream.
8230 @subsubsection Hald CLUT with preview
8232 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
8233 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
8234 biggest possible square starting at the top left of the picture. The remaining
8235 padding pixels (bottom or right) will be ignored. This area can be used to add
8236 a preview of the Hald CLUT.
8238 Typically, the following generated Hald CLUT will be supported by the
8239 @code{haldclut} filter:
8242 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
8243 pad=iw+320 [padded_clut];
8244 smptebars=s=320x256, split [a][b];
8245 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
8246 [main][b] overlay=W-320" -frames:v 1 clut.png
8249 It contains the original and a preview of the effect of the CLUT: SMPTE color
8250 bars are displayed on the right-top, and below the same color bars processed by
8253 Then, the effect of this Hald CLUT can be visualized with:
8255 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
8260 Flip the input video horizontally.
8262 For example, to horizontally flip the input video with @command{ffmpeg}:
8264 ffmpeg -i in.avi -vf "hflip" out.avi
8268 This filter applies a global color histogram equalization on a
8271 It can be used to correct video that has a compressed range of pixel
8272 intensities. The filter redistributes the pixel intensities to
8273 equalize their distribution across the intensity range. It may be
8274 viewed as an "automatically adjusting contrast filter". This filter is
8275 useful only for correcting degraded or poorly captured source
8278 The filter accepts the following options:
8282 Determine the amount of equalization to be applied. As the strength
8283 is reduced, the distribution of pixel intensities more-and-more
8284 approaches that of the input frame. The value must be a float number
8285 in the range [0,1] and defaults to 0.200.
8288 Set the maximum intensity that can generated and scale the output
8289 values appropriately. The strength should be set as desired and then
8290 the intensity can be limited if needed to avoid washing-out. The value
8291 must be a float number in the range [0,1] and defaults to 0.210.
8294 Set the antibanding level. If enabled the filter will randomly vary
8295 the luminance of output pixels by a small amount to avoid banding of
8296 the histogram. Possible values are @code{none}, @code{weak} or
8297 @code{strong}. It defaults to @code{none}.
8302 Compute and draw a color distribution histogram for the input video.
8304 The computed histogram is a representation of the color component
8305 distribution in an image.
8307 Standard histogram displays the color components distribution in an image.
8308 Displays color graph for each color component. Shows distribution of
8309 the Y, U, V, A or R, G, B components, depending on input format, in the
8310 current frame. Below each graph a color component scale meter is shown.
8312 The filter accepts the following options:
8316 Set height of level. Default value is @code{200}.
8317 Allowed range is [50, 2048].
8320 Set height of color scale. Default value is @code{12}.
8321 Allowed range is [0, 40].
8325 It accepts the following values:
8328 Per color component graphs are placed below each other.
8331 Presents information identical to that in the @code{parade}, except
8332 that the graphs representing color components are superimposed directly
8335 Default is @code{parade}.
8338 Set mode. Can be either @code{linear}, or @code{logarithmic}.
8339 Default is @code{linear}.
8342 Set what color components to display.
8343 Default is @code{7}.
8346 @subsection Examples
8351 Calculate and draw histogram:
8353 ffplay -i input -vf histogram
8361 This is a high precision/quality 3d denoise filter. It aims to reduce
8362 image noise, producing smooth images and making still images really
8363 still. It should enhance compressibility.
8365 It accepts the following optional parameters:
8369 A non-negative floating point number which specifies spatial luma strength.
8372 @item chroma_spatial
8373 A non-negative floating point number which specifies spatial chroma strength.
8374 It defaults to 3.0*@var{luma_spatial}/4.0.
8377 A floating point number which specifies luma temporal strength. It defaults to
8378 6.0*@var{luma_spatial}/4.0.
8381 A floating point number which specifies chroma temporal strength. It defaults to
8382 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}.
8385 @anchor{hwupload_cuda}
8386 @section hwupload_cuda
8388 Upload system memory frames to a CUDA device.
8390 It accepts the following optional parameters:
8394 The number of the CUDA device to use
8399 Apply a high-quality magnification filter designed for pixel art. This filter
8400 was originally created by Maxim Stepin.
8402 It accepts the following option:
8406 Set the scaling dimension: @code{2} for @code{hq2x}, @code{3} for
8407 @code{hq3x} and @code{4} for @code{hq4x}.
8408 Default is @code{3}.
8412 Stack input videos horizontally.
8414 All streams must be of same pixel format and of same height.
8416 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
8417 to create same output.
8419 The filter accept the following option:
8423 Set number of input streams. Default is 2.
8426 If set to 1, force the output to terminate when the shortest input
8427 terminates. Default value is 0.
8432 Modify the hue and/or the saturation of the input.
8434 It accepts the following parameters:
8438 Specify the hue angle as a number of degrees. It accepts an expression,
8439 and defaults to "0".
8442 Specify the saturation in the [-10,10] range. It accepts an expression and
8446 Specify the hue angle as a number of radians. It accepts an
8447 expression, and defaults to "0".
8450 Specify the brightness in the [-10,10] range. It accepts an expression and
8454 @option{h} and @option{H} are mutually exclusive, and can't be
8455 specified at the same time.
8457 The @option{b}, @option{h}, @option{H} and @option{s} option values are
8458 expressions containing the following constants:
8462 frame count of the input frame starting from 0
8465 presentation timestamp of the input frame expressed in time base units
8468 frame rate of the input video, NAN if the input frame rate is unknown
8471 timestamp expressed in seconds, NAN if the input timestamp is unknown
8474 time base of the input video
8477 @subsection Examples
8481 Set the hue to 90 degrees and the saturation to 1.0:
8487 Same command but expressing the hue in radians:
8493 Rotate hue and make the saturation swing between 0
8494 and 2 over a period of 1 second:
8496 hue="H=2*PI*t: s=sin(2*PI*t)+1"
8500 Apply a 3 seconds saturation fade-in effect starting at 0:
8505 The general fade-in expression can be written as:
8507 hue="s=min(0\, max((t-START)/DURATION\, 1))"
8511 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
8513 hue="s=max(0\, min(1\, (8-t)/3))"
8516 The general fade-out expression can be written as:
8518 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
8523 @subsection Commands
8525 This filter supports the following commands:
8531 Modify the hue and/or the saturation and/or brightness of the input video.
8532 The command accepts the same syntax of the corresponding option.
8534 If the specified expression is not valid, it is kept at its current
8540 Detect video interlacing type.
8542 This filter tries to detect if the input frames as interlaced, progressive,
8543 top or bottom field first. It will also try and detect fields that are
8544 repeated between adjacent frames (a sign of telecine).
8546 Single frame detection considers only immediately adjacent frames when classifying each frame.
8547 Multiple frame detection incorporates the classification history of previous frames.
8549 The filter will log these metadata values:
8552 @item single.current_frame
8553 Detected type of current frame using single-frame detection. One of:
8554 ``tff'' (top field first), ``bff'' (bottom field first),
8555 ``progressive'', or ``undetermined''
8558 Cumulative number of frames detected as top field first using single-frame detection.
8561 Cumulative number of frames detected as top field first using multiple-frame detection.
8564 Cumulative number of frames detected as bottom field first using single-frame detection.
8566 @item multiple.current_frame
8567 Detected type of current frame using multiple-frame detection. One of:
8568 ``tff'' (top field first), ``bff'' (bottom field first),
8569 ``progressive'', or ``undetermined''
8572 Cumulative number of frames detected as bottom field first using multiple-frame detection.
8574 @item single.progressive
8575 Cumulative number of frames detected as progressive using single-frame detection.
8577 @item multiple.progressive
8578 Cumulative number of frames detected as progressive using multiple-frame detection.
8580 @item single.undetermined
8581 Cumulative number of frames that could not be classified using single-frame detection.
8583 @item multiple.undetermined
8584 Cumulative number of frames that could not be classified using multiple-frame detection.
8586 @item repeated.current_frame
8587 Which field in the current frame is repeated from the last. One of ``neither'', ``top'', or ``bottom''.
8589 @item repeated.neither
8590 Cumulative number of frames with no repeated field.
8593 Cumulative number of frames with the top field repeated from the previous frame's top field.
8595 @item repeated.bottom
8596 Cumulative number of frames with the bottom field repeated from the previous frame's bottom field.
8599 The filter accepts the following options:
8603 Set interlacing threshold.
8605 Set progressive threshold.
8607 Threshold for repeated field detection.
8609 Number of frames after which a given frame's contribution to the
8610 statistics is halved (i.e., it contributes only 0.5 to it's
8611 classification). The default of 0 means that all frames seen are given
8612 full weight of 1.0 forever.
8613 @item analyze_interlaced_flag
8614 When this is not 0 then idet will use the specified number of frames to determine
8615 if the interlaced flag is accurate, it will not count undetermined frames.
8616 If the flag is found to be accurate it will be used without any further
8617 computations, if it is found to be inaccurate it will be cleared without any
8618 further computations. This allows inserting the idet filter as a low computational
8619 method to clean up the interlaced flag
8624 Deinterleave or interleave fields.
8626 This filter allows one to process interlaced images fields without
8627 deinterlacing them. Deinterleaving splits the input frame into 2
8628 fields (so called half pictures). Odd lines are moved to the top
8629 half of the output image, even lines to the bottom half.
8630 You can process (filter) them independently and then re-interleave them.
8632 The filter accepts the following options:
8636 @item chroma_mode, c
8638 Available values for @var{luma_mode}, @var{chroma_mode} and
8639 @var{alpha_mode} are:
8645 @item deinterleave, d
8646 Deinterleave fields, placing one above the other.
8649 Interleave fields. Reverse the effect of deinterleaving.
8651 Default value is @code{none}.
8654 @item chroma_swap, cs
8655 @item alpha_swap, as
8656 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
8661 Apply inflate effect to the video.
8663 This filter replaces the pixel by the local(3x3) average by taking into account
8664 only values higher than the pixel.
8666 It accepts the following options:
8673 Limit the maximum change for each plane, default is 65535.
8674 If 0, plane will remain unchanged.
8679 Simple interlacing filter from progressive contents. This interleaves upper (or
8680 lower) lines from odd frames with lower (or upper) lines from even frames,
8681 halving the frame rate and preserving image height.
8684 Original Original New Frame
8685 Frame 'j' Frame 'j+1' (tff)
8686 ========== =========== ==================
8687 Line 0 --------------------> Frame 'j' Line 0
8688 Line 1 Line 1 ----> Frame 'j+1' Line 1
8689 Line 2 ---------------------> Frame 'j' Line 2
8690 Line 3 Line 3 ----> Frame 'j+1' Line 3
8692 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
8695 It accepts the following optional parameters:
8699 This determines whether the interlaced frame is taken from the even
8700 (tff - default) or odd (bff) lines of the progressive frame.
8703 Enable (default) or disable the vertical lowpass filter to avoid twitter
8704 interlacing and reduce moire patterns.
8709 Deinterlace input video by applying Donald Graft's adaptive kernel
8710 deinterling. Work on interlaced parts of a video to produce
8713 The description of the accepted parameters follows.
8717 Set the threshold which affects the filter's tolerance when
8718 determining if a pixel line must be processed. It must be an integer
8719 in the range [0,255] and defaults to 10. A value of 0 will result in
8720 applying the process on every pixels.
8723 Paint pixels exceeding the threshold value to white if set to 1.
8727 Set the fields order. Swap fields if set to 1, leave fields alone if
8731 Enable additional sharpening if set to 1. Default is 0.
8734 Enable twoway sharpening if set to 1. Default is 0.
8737 @subsection Examples
8741 Apply default values:
8743 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
8747 Enable additional sharpening:
8753 Paint processed pixels in white:
8759 @section lenscorrection
8761 Correct radial lens distortion
8763 This filter can be used to correct for radial distortion as can result from the use
8764 of wide angle lenses, and thereby re-rectify the image. To find the right parameters
8765 one can use tools available for example as part of opencv or simply trial-and-error.
8766 To use opencv use the calibration sample (under samples/cpp) from the opencv sources
8767 and extract the k1 and k2 coefficients from the resulting matrix.
8769 Note that effectively the same filter is available in the open-source tools Krita and
8770 Digikam from the KDE project.
8772 In contrast to the @ref{vignette} filter, which can also be used to compensate lens errors,
8773 this filter corrects the distortion of the image, whereas @ref{vignette} corrects the
8774 brightness distribution, so you may want to use both filters together in certain
8775 cases, though you will have to take care of ordering, i.e. whether vignetting should
8776 be applied before or after lens correction.
8780 The filter accepts the following options:
8784 Relative x-coordinate of the focal point of the image, and thereby the center of the
8785 distortion. This value has a range [0,1] and is expressed as fractions of the image
8788 Relative y-coordinate of the focal point of the image, and thereby the center of the
8789 distortion. This value has a range [0,1] and is expressed as fractions of the image
8792 Coefficient of the quadratic correction term. 0.5 means no correction.
8794 Coefficient of the double quadratic correction term. 0.5 means no correction.
8797 The formula that generates the correction is:
8799 @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)
8801 where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the
8802 distances from the focal point in the source and target images, respectively.
8804 @section loop, aloop
8806 Loop video frames or audio samples.
8808 Those filters accepts the following options:
8812 Set the number of loops.
8815 Set maximal size in number of frames for @code{loop} filter or maximal number
8816 of samples in case of @code{aloop} filter.
8819 Set first frame of loop for @code{loop} filter or first sample of loop in case
8820 of @code{aloop} filter.
8826 Apply a 3D LUT to an input video.
8828 The filter accepts the following options:
8832 Set the 3D LUT file name.
8834 Currently supported formats:
8846 Select interpolation mode.
8848 Available values are:
8852 Use values from the nearest defined point.
8854 Interpolate values using the 8 points defining a cube.
8856 Interpolate values using a tetrahedron.
8860 @section lut, lutrgb, lutyuv
8862 Compute a look-up table for binding each pixel component input value
8863 to an output value, and apply it to the input video.
8865 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
8866 to an RGB input video.
8868 These filters accept the following parameters:
8871 set first pixel component expression
8873 set second pixel component expression
8875 set third pixel component expression
8877 set fourth pixel component expression, corresponds to the alpha component
8880 set red component expression
8882 set green component expression
8884 set blue component expression
8886 alpha component expression
8889 set Y/luminance component expression
8891 set U/Cb component expression
8893 set V/Cr component expression
8896 Each of them specifies the expression to use for computing the lookup table for
8897 the corresponding pixel component values.
8899 The exact component associated to each of the @var{c*} options depends on the
8902 The @var{lut} filter requires either YUV or RGB pixel formats in input,
8903 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
8905 The expressions can contain the following constants and functions:
8910 The input width and height.
8913 The input value for the pixel component.
8916 The input value, clipped to the @var{minval}-@var{maxval} range.
8919 The maximum value for the pixel component.
8922 The minimum value for the pixel component.
8925 The negated value for the pixel component value, clipped to the
8926 @var{minval}-@var{maxval} range; it corresponds to the expression
8927 "maxval-clipval+minval".
8930 The computed value in @var{val}, clipped to the
8931 @var{minval}-@var{maxval} range.
8933 @item gammaval(gamma)
8934 The computed gamma correction value of the pixel component value,
8935 clipped to the @var{minval}-@var{maxval} range. It corresponds to the
8937 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
8941 All expressions default to "val".
8943 @subsection Examples
8949 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
8950 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
8953 The above is the same as:
8955 lutrgb="r=negval:g=negval:b=negval"
8956 lutyuv="y=negval:u=negval:v=negval"
8966 Remove chroma components, turning the video into a graytone image:
8968 lutyuv="u=128:v=128"
8972 Apply a luma burning effect:
8978 Remove green and blue components:
8984 Set a constant alpha channel value on input:
8986 format=rgba,lutrgb=a="maxval-minval/2"
8990 Correct luminance gamma by a factor of 0.5:
8992 lutyuv=y=gammaval(0.5)
8996 Discard least significant bits of luma:
8998 lutyuv=y='bitand(val, 128+64+32)'
9002 @section maskedmerge
9004 Merge the first input stream with the second input stream using per pixel
9005 weights in the third input stream.
9007 A value of 0 in the third stream pixel component means that pixel component
9008 from first stream is returned unchanged, while maximum value (eg. 255 for
9009 8-bit videos) means that pixel component from second stream is returned
9010 unchanged. Intermediate values define the amount of merging between both
9011 input stream's pixel components.
9013 This filter accepts the following options:
9016 Set which planes will be processed as bitmap, unprocessed planes will be
9017 copied from first stream.
9018 By default value 0xf, all planes will be processed.
9023 Apply motion-compensation deinterlacing.
9025 It needs one field per frame as input and must thus be used together
9026 with yadif=1/3 or equivalent.
9028 This filter accepts the following options:
9031 Set the deinterlacing mode.
9033 It accepts one of the following values:
9038 use iterative motion estimation
9040 like @samp{slow}, but use multiple reference frames.
9042 Default value is @samp{fast}.
9045 Set the picture field parity assumed for the input video. It must be
9046 one of the following values:
9050 assume top field first
9052 assume bottom field first
9055 Default value is @samp{bff}.
9058 Set per-block quantization parameter (QP) used by the internal
9061 Higher values should result in a smoother motion vector field but less
9062 optimal individual vectors. Default value is 1.
9065 @section mergeplanes
9067 Merge color channel components from several video streams.
9069 The filter accepts up to 4 input streams, and merge selected input
9070 planes to the output video.
9072 This filter accepts the following options:
9075 Set input to output plane mapping. Default is @code{0}.
9077 The mappings is specified as a bitmap. It should be specified as a
9078 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
9079 mapping for the first plane of the output stream. 'A' sets the number of
9080 the input stream to use (from 0 to 3), and 'a' the plane number of the
9081 corresponding input to use (from 0 to 3). The rest of the mappings is
9082 similar, 'Bb' describes the mapping for the output stream second
9083 plane, 'Cc' describes the mapping for the output stream third plane and
9084 'Dd' describes the mapping for the output stream fourth plane.
9087 Set output pixel format. Default is @code{yuva444p}.
9090 @subsection Examples
9094 Merge three gray video streams of same width and height into single video stream:
9096 [a0][a1][a2]mergeplanes=0x001020:yuv444p
9100 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
9102 [a0][a1]mergeplanes=0x00010210:yuva444p
9106 Swap Y and A plane in yuva444p stream:
9108 format=yuva444p,mergeplanes=0x03010200:yuva444p
9112 Swap U and V plane in yuv420p stream:
9114 format=yuv420p,mergeplanes=0x000201:yuv420p
9118 Cast a rgb24 clip to yuv444p:
9120 format=rgb24,mergeplanes=0x000102:yuv444p
9124 @section metadata, ametadata
9126 Manipulate frame metadata.
9128 This filter accepts the following options:
9132 Set mode of operation of the filter.
9134 Can be one of the following:
9138 If both @code{value} and @code{key} is set, select frames
9139 which have such metadata. If only @code{key} is set, select
9140 every frame that has such key in metadata.
9143 Add new metadata @code{key} and @code{value}. If key is already available
9147 Modify value of already present key.
9150 If @code{value} is set, delete only keys that have such value.
9151 Otherwise, delete key.
9154 Print key and its value if metadata was found. If @code{key} is not set print all
9155 metadata values available in frame.
9159 Set key used with all modes. Must be set for all modes except @code{print}.
9162 Set metadata value which will be used. This option is mandatory for
9163 @code{modify} and @code{add} mode.
9166 Which function to use when comparing metadata value and @code{value}.
9168 Can be one of following:
9172 Values are interpreted as strings, returns true if metadata value is same as @code{value}.
9175 Values are interpreted as strings, returns true if metadata value starts with
9176 the @code{value} option string.
9179 Values are interpreted as floats, returns true if metadata value is less than @code{value}.
9182 Values are interpreted as floats, returns true if @code{value} is equal with metadata value.
9185 Values are interpreted as floats, returns true if metadata value is greater than @code{value}.
9188 Values are interpreted as floats, returns true if expression from option @code{expr}
9193 Set expression which is used when @code{function} is set to @code{expr}.
9194 The expression is evaluated through the eval API and can contain the following
9199 Float representation of @code{value} from metadata key.
9202 Float representation of @code{value} as supplied by user in @code{value} option.
9206 If specified in @code{print} mode, output is written to the named file. When
9207 filename equals "-" data is written to standard output.
9208 If @code{file} option is not set, output is written to the log with AV_LOG_INFO
9212 @subsection Examples
9216 Print all metadata values for frames with key @code{lavfi.singnalstats.YDIF} with values
9220 signalstats,metadata=print:key=lavfi.signalstats.YDIF:value=0:function=expr:expr='between(VALUE1,0,1)'
9225 Drop frames that do not differ greatly from the previous frame in
9226 order to reduce frame rate.
9228 The main use of this filter is for very-low-bitrate encoding
9229 (e.g. streaming over dialup modem), but it could in theory be used for
9230 fixing movies that were inverse-telecined incorrectly.
9232 A description of the accepted options follows.
9236 Set the maximum number of consecutive frames which can be dropped (if
9237 positive), or the minimum interval between dropped frames (if
9238 negative). If the value is 0, the frame is dropped unregarding the
9239 number of previous sequentially dropped frames.
9246 Set the dropping threshold values.
9248 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
9249 represent actual pixel value differences, so a threshold of 64
9250 corresponds to 1 unit of difference for each pixel, or the same spread
9251 out differently over the block.
9253 A frame is a candidate for dropping if no 8x8 blocks differ by more
9254 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
9255 meaning the whole image) differ by more than a threshold of @option{lo}.
9257 Default value for @option{hi} is 64*12, default value for @option{lo} is
9258 64*5, and default value for @option{frac} is 0.33.
9266 It accepts an integer in input; if non-zero it negates the
9267 alpha component (if available). The default value in input is 0.
9271 Deinterlace video using neural network edge directed interpolation.
9273 This filter accepts the following options:
9277 Mandatory option, without binary file filter can not work.
9278 Currently file can be found here:
9279 https://github.com/dubhater/vapoursynth-nnedi3/blob/master/src/nnedi3_weights.bin
9282 Set which frames to deinterlace, by default it is @code{all}.
9283 Can be @code{all} or @code{interlaced}.
9286 Set mode of operation.
9288 Can be one of the following:
9292 Use frame flags, both fields.
9294 Use frame flags, single field.
9298 Use bottom field only.
9300 Use both fields, top first.
9302 Use both fields, bottom first.
9306 Set which planes to process, by default filter process all frames.
9309 Set size of local neighborhood around each pixel, used by the predictor neural
9312 Can be one of the following:
9325 Set the number of neurons in predicctor neural network.
9326 Can be one of the following:
9337 Controls the number of different neural network predictions that are blended
9338 together to compute the final output value. Can be @code{fast}, default or
9342 Set which set of weights to use in the predictor.
9343 Can be one of the following:
9347 weights trained to minimize absolute error
9349 weights trained to minimize squared error
9353 Controls whether or not the prescreener neural network is used to decide
9354 which pixels should be processed by the predictor neural network and which
9355 can be handled by simple cubic interpolation.
9356 The prescreener is trained to know whether cubic interpolation will be
9357 sufficient for a pixel or whether it should be predicted by the predictor nn.
9358 The computational complexity of the prescreener nn is much less than that of
9359 the predictor nn. Since most pixels can be handled by cubic interpolation,
9360 using the prescreener generally results in much faster processing.
9361 The prescreener is pretty accurate, so the difference between using it and not
9362 using it is almost always unnoticeable.
9364 Can be one of the following:
9372 Default is @code{new}.
9375 Set various debugging flags.
9380 Force libavfilter not to use any of the specified pixel formats for the
9381 input to the next filter.
9383 It accepts the following parameters:
9387 A '|'-separated list of pixel format names, such as
9388 apix_fmts=yuv420p|monow|rgb24".
9392 @subsection Examples
9396 Force libavfilter to use a format different from @var{yuv420p} for the
9397 input to the vflip filter:
9399 noformat=pix_fmts=yuv420p,vflip
9403 Convert the input video to any of the formats not contained in the list:
9405 noformat=yuv420p|yuv444p|yuv410p
9411 Add noise on video input frame.
9413 The filter accepts the following options:
9421 Set noise seed for specific pixel component or all pixel components in case
9422 of @var{all_seed}. Default value is @code{123457}.
9424 @item all_strength, alls
9425 @item c0_strength, c0s
9426 @item c1_strength, c1s
9427 @item c2_strength, c2s
9428 @item c3_strength, c3s
9429 Set noise strength for specific pixel component or all pixel components in case
9430 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
9432 @item all_flags, allf
9437 Set pixel component flags or set flags for all components if @var{all_flags}.
9438 Available values for component flags are:
9441 averaged temporal noise (smoother)
9443 mix random noise with a (semi)regular pattern
9445 temporal noise (noise pattern changes between frames)
9447 uniform noise (gaussian otherwise)
9451 @subsection Examples
9453 Add temporal and uniform noise to input video:
9455 noise=alls=20:allf=t+u
9460 Pass the video source unchanged to the output.
9463 Optical Character Recognition
9465 This filter uses Tesseract for optical character recognition.
9467 It accepts the following options:
9471 Set datapath to tesseract data. Default is to use whatever was
9472 set at installation.
9475 Set language, default is "eng".
9478 Set character whitelist.
9481 Set character blacklist.
9484 The filter exports recognized text as the frame metadata @code{lavfi.ocr.text}.
9488 Apply a video transform using libopencv.
9490 To enable this filter, install the libopencv library and headers and
9491 configure FFmpeg with @code{--enable-libopencv}.
9493 It accepts the following parameters:
9498 The name of the libopencv filter to apply.
9501 The parameters to pass to the libopencv filter. If not specified, the default
9506 Refer to the official libopencv documentation for more precise
9508 @url{http://docs.opencv.org/master/modules/imgproc/doc/filtering.html}
9510 Several libopencv filters are supported; see the following subsections.
9515 Dilate an image by using a specific structuring element.
9516 It corresponds to the libopencv function @code{cvDilate}.
9518 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
9520 @var{struct_el} represents a structuring element, and has the syntax:
9521 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
9523 @var{cols} and @var{rows} represent the number of columns and rows of
9524 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
9525 point, and @var{shape} the shape for the structuring element. @var{shape}
9526 must be "rect", "cross", "ellipse", or "custom".
9528 If the value for @var{shape} is "custom", it must be followed by a
9529 string of the form "=@var{filename}". The file with name
9530 @var{filename} is assumed to represent a binary image, with each
9531 printable character corresponding to a bright pixel. When a custom
9532 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
9533 or columns and rows of the read file are assumed instead.
9535 The default value for @var{struct_el} is "3x3+0x0/rect".
9537 @var{nb_iterations} specifies the number of times the transform is
9538 applied to the image, and defaults to 1.
9542 # Use the default values
9545 # Dilate using a structuring element with a 5x5 cross, iterating two times
9546 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
9548 # Read the shape from the file diamond.shape, iterating two times.
9549 # The file diamond.shape may contain a pattern of characters like this
9555 # The specified columns and rows are ignored
9556 # but the anchor point coordinates are not
9557 ocv=dilate:0x0+2x2/custom=diamond.shape|2
9562 Erode an image by using a specific structuring element.
9563 It corresponds to the libopencv function @code{cvErode}.
9565 It accepts the parameters: @var{struct_el}:@var{nb_iterations},
9566 with the same syntax and semantics as the @ref{dilate} filter.
9570 Smooth the input video.
9572 The filter takes the following parameters:
9573 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
9575 @var{type} is the type of smooth filter to apply, and must be one of
9576 the following values: "blur", "blur_no_scale", "median", "gaussian",
9577 or "bilateral". The default value is "gaussian".
9579 The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
9580 depend on the smooth type. @var{param1} and
9581 @var{param2} accept integer positive values or 0. @var{param3} and
9582 @var{param4} accept floating point values.
9584 The default value for @var{param1} is 3. The default value for the
9585 other parameters is 0.
9587 These parameters correspond to the parameters assigned to the
9588 libopencv function @code{cvSmooth}.
9593 Overlay one video on top of another.
9595 It takes two inputs and has one output. The first input is the "main"
9596 video on which the second input is overlaid.
9598 It accepts the following parameters:
9600 A description of the accepted options follows.
9605 Set the expression for the x and y coordinates of the overlaid video
9606 on the main video. Default value is "0" for both expressions. In case
9607 the expression is invalid, it is set to a huge value (meaning that the
9608 overlay will not be displayed within the output visible area).
9611 The action to take when EOF is encountered on the secondary input; it accepts
9612 one of the following values:
9616 Repeat the last frame (the default).
9620 Pass the main input through.
9624 Set when the expressions for @option{x}, and @option{y} are evaluated.
9626 It accepts the following values:
9629 only evaluate expressions once during the filter initialization or
9630 when a command is processed
9633 evaluate expressions for each incoming frame
9636 Default value is @samp{frame}.
9639 If set to 1, force the output to terminate when the shortest input
9640 terminates. Default value is 0.
9643 Set the format for the output video.
9645 It accepts the following values:
9660 Default value is @samp{yuv420}.
9662 @item rgb @emph{(deprecated)}
9663 If set to 1, force the filter to accept inputs in the RGB
9664 color space. Default value is 0. This option is deprecated, use
9665 @option{format} instead.
9668 If set to 1, force the filter to draw the last overlay frame over the
9669 main input until the end of the stream. A value of 0 disables this
9670 behavior. Default value is 1.
9673 The @option{x}, and @option{y} expressions can contain the following
9679 The main input width and height.
9683 The overlay input width and height.
9687 The computed values for @var{x} and @var{y}. They are evaluated for
9692 horizontal and vertical chroma subsample values of the output
9693 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
9697 the number of input frame, starting from 0
9700 the position in the file of the input frame, NAN if unknown
9703 The timestamp, expressed in seconds. It's NAN if the input timestamp is unknown.
9707 Note that the @var{n}, @var{pos}, @var{t} variables are available only
9708 when evaluation is done @emph{per frame}, and will evaluate to NAN
9709 when @option{eval} is set to @samp{init}.
9711 Be aware that frames are taken from each input video in timestamp
9712 order, hence, if their initial timestamps differ, it is a good idea
9713 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
9714 have them begin in the same zero timestamp, as the example for
9715 the @var{movie} filter does.
9717 You can chain together more overlays but you should test the
9718 efficiency of such approach.
9720 @subsection Commands
9722 This filter supports the following commands:
9726 Modify the x and y of the overlay input.
9727 The command accepts the same syntax of the corresponding option.
9729 If the specified expression is not valid, it is kept at its current
9733 @subsection Examples
9737 Draw the overlay at 10 pixels from the bottom right corner of the main
9740 overlay=main_w-overlay_w-10:main_h-overlay_h-10
9743 Using named options the example above becomes:
9745 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
9749 Insert a transparent PNG logo in the bottom left corner of the input,
9750 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
9752 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
9756 Insert 2 different transparent PNG logos (second logo on bottom
9757 right corner) using the @command{ffmpeg} tool:
9759 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
9763 Add a transparent color layer on top of the main video; @code{WxH}
9764 must specify the size of the main input to the overlay filter:
9766 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
9770 Play an original video and a filtered version (here with the deshake
9771 filter) side by side using the @command{ffplay} tool:
9773 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
9776 The above command is the same as:
9778 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
9782 Make a sliding overlay appearing from the left to the right top part of the
9783 screen starting since time 2:
9785 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
9789 Compose output by putting two input videos side to side:
9791 ffmpeg -i left.avi -i right.avi -filter_complex "
9792 nullsrc=size=200x100 [background];
9793 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
9794 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
9795 [background][left] overlay=shortest=1 [background+left];
9796 [background+left][right] overlay=shortest=1:x=100 [left+right]
9801 Mask 10-20 seconds of a video by applying the delogo filter to a section
9803 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
9804 -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]'
9809 Chain several overlays in cascade:
9811 nullsrc=s=200x200 [bg];
9812 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
9813 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
9814 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
9815 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
9816 [in3] null, [mid2] overlay=100:100 [out0]
9823 Apply Overcomplete Wavelet denoiser.
9825 The filter accepts the following options:
9831 Larger depth values will denoise lower frequency components more, but
9832 slow down filtering.
9834 Must be an int in the range 8-16, default is @code{8}.
9836 @item luma_strength, ls
9839 Must be a double value in the range 0-1000, default is @code{1.0}.
9841 @item chroma_strength, cs
9842 Set chroma strength.
9844 Must be a double value in the range 0-1000, default is @code{1.0}.
9850 Add paddings to the input image, and place the original input at the
9851 provided @var{x}, @var{y} coordinates.
9853 It accepts the following parameters:
9858 Specify an expression for the size of the output image with the
9859 paddings added. If the value for @var{width} or @var{height} is 0, the
9860 corresponding input size is used for the output.
9862 The @var{width} expression can reference the value set by the
9863 @var{height} expression, and vice versa.
9865 The default value of @var{width} and @var{height} is 0.
9869 Specify the offsets to place the input image at within the padded area,
9870 with respect to the top/left border of the output image.
9872 The @var{x} expression can reference the value set by the @var{y}
9873 expression, and vice versa.
9875 The default value of @var{x} and @var{y} is 0.
9878 Specify the color of the padded area. For the syntax of this option,
9879 check the "Color" section in the ffmpeg-utils manual.
9881 The default value of @var{color} is "black".
9884 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
9885 options are expressions containing the following constants:
9890 The input video width and height.
9894 These are the same as @var{in_w} and @var{in_h}.
9898 The output width and height (the size of the padded area), as
9899 specified by the @var{width} and @var{height} expressions.
9903 These are the same as @var{out_w} and @var{out_h}.
9907 The x and y offsets as specified by the @var{x} and @var{y}
9908 expressions, or NAN if not yet specified.
9911 same as @var{iw} / @var{ih}
9914 input sample aspect ratio
9917 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
9921 The horizontal and vertical chroma subsample values. For example for the
9922 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
9925 @subsection Examples
9929 Add paddings with the color "violet" to the input video. The output video
9930 size is 640x480, and the top-left corner of the input video is placed at
9933 pad=640:480:0:40:violet
9936 The example above is equivalent to the following command:
9938 pad=width=640:height=480:x=0:y=40:color=violet
9942 Pad the input to get an output with dimensions increased by 3/2,
9943 and put the input video at the center of the padded area:
9945 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
9949 Pad the input to get a squared output with size equal to the maximum
9950 value between the input width and height, and put the input video at
9951 the center of the padded area:
9953 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
9957 Pad the input to get a final w/h ratio of 16:9:
9959 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
9963 In case of anamorphic video, in order to set the output display aspect
9964 correctly, it is necessary to use @var{sar} in the expression,
9965 according to the relation:
9967 (ih * X / ih) * sar = output_dar
9968 X = output_dar / sar
9971 Thus the previous example needs to be modified to:
9973 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
9977 Double the output size and put the input video in the bottom-right
9978 corner of the output padded area:
9980 pad="2*iw:2*ih:ow-iw:oh-ih"
9987 Generate one palette for a whole video stream.
9989 It accepts the following options:
9993 Set the maximum number of colors to quantize in the palette.
9994 Note: the palette will still contain 256 colors; the unused palette entries
9997 @item reserve_transparent
9998 Create a palette of 255 colors maximum and reserve the last one for
9999 transparency. Reserving the transparency color is useful for GIF optimization.
10000 If not set, the maximum of colors in the palette will be 256. You probably want
10001 to disable this option for a standalone image.
10005 Set statistics mode.
10007 It accepts the following values:
10010 Compute full frame histograms.
10012 Compute histograms only for the part that differs from previous frame. This
10013 might be relevant to give more importance to the moving part of your input if
10014 the background is static.
10017 Default value is @var{full}.
10020 The filter also exports the frame metadata @code{lavfi.color_quant_ratio}
10021 (@code{nb_color_in / nb_color_out}) which you can use to evaluate the degree of
10022 color quantization of the palette. This information is also visible at
10023 @var{info} logging level.
10025 @subsection Examples
10029 Generate a representative palette of a given video using @command{ffmpeg}:
10031 ffmpeg -i input.mkv -vf palettegen palette.png
10035 @section paletteuse
10037 Use a palette to downsample an input video stream.
10039 The filter takes two inputs: one video stream and a palette. The palette must
10040 be a 256 pixels image.
10042 It accepts the following options:
10046 Select dithering mode. Available algorithms are:
10049 Ordered 8x8 bayer dithering (deterministic)
10051 Dithering as defined by Paul Heckbert in 1982 (simple error diffusion).
10052 Note: this dithering is sometimes considered "wrong" and is included as a
10054 @item floyd_steinberg
10055 Floyd and Steingberg dithering (error diffusion)
10057 Frankie Sierra dithering v2 (error diffusion)
10059 Frankie Sierra dithering v2 "Lite" (error diffusion)
10062 Default is @var{sierra2_4a}.
10065 When @var{bayer} dithering is selected, this option defines the scale of the
10066 pattern (how much the crosshatch pattern is visible). A low value means more
10067 visible pattern for less banding, and higher value means less visible pattern
10068 at the cost of more banding.
10070 The option must be an integer value in the range [0,5]. Default is @var{2}.
10073 If set, define the zone to process
10077 Only the changing rectangle will be reprocessed. This is similar to GIF
10078 cropping/offsetting compression mechanism. This option can be useful for speed
10079 if only a part of the image is changing, and has use cases such as limiting the
10080 scope of the error diffusal @option{dither} to the rectangle that bounds the
10081 moving scene (it leads to more deterministic output if the scene doesn't change
10082 much, and as a result less moving noise and better GIF compression).
10085 Default is @var{none}.
10088 @subsection Examples
10092 Use a palette (generated for example with @ref{palettegen}) to encode a GIF
10093 using @command{ffmpeg}:
10095 ffmpeg -i input.mkv -i palette.png -lavfi paletteuse output.gif
10099 @section perspective
10101 Correct perspective of video not recorded perpendicular to the screen.
10103 A description of the accepted parameters follows.
10114 Set coordinates expression for top left, top right, bottom left and bottom right corners.
10115 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
10116 If the @code{sense} option is set to @code{source}, then the specified points will be sent
10117 to the corners of the destination. If the @code{sense} option is set to @code{destination},
10118 then the corners of the source will be sent to the specified coordinates.
10120 The expressions can use the following variables:
10125 the width and height of video frame.
10128 @item interpolation
10129 Set interpolation for perspective correction.
10131 It accepts the following values:
10137 Default value is @samp{linear}.
10140 Set interpretation of coordinate options.
10142 It accepts the following values:
10146 Send point in the source specified by the given coordinates to
10147 the corners of the destination.
10149 @item 1, destination
10151 Send the corners of the source to the point in the destination specified
10152 by the given coordinates.
10154 Default value is @samp{source}.
10160 Delay interlaced video by one field time so that the field order changes.
10162 The intended use is to fix PAL movies that have been captured with the
10163 opposite field order to the film-to-video transfer.
10165 A description of the accepted parameters follows.
10171 It accepts the following values:
10174 Capture field order top-first, transfer bottom-first.
10175 Filter will delay the bottom field.
10178 Capture field order bottom-first, transfer top-first.
10179 Filter will delay the top field.
10182 Capture and transfer with the same field order. This mode only exists
10183 for the documentation of the other options to refer to, but if you
10184 actually select it, the filter will faithfully do nothing.
10187 Capture field order determined automatically by field flags, transfer
10189 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
10190 basis using field flags. If no field information is available,
10191 then this works just like @samp{u}.
10194 Capture unknown or varying, transfer opposite.
10195 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
10196 analyzing the images and selecting the alternative that produces best
10197 match between the fields.
10200 Capture top-first, transfer unknown or varying.
10201 Filter selects among @samp{t} and @samp{p} using image analysis.
10204 Capture bottom-first, transfer unknown or varying.
10205 Filter selects among @samp{b} and @samp{p} using image analysis.
10208 Capture determined by field flags, transfer unknown or varying.
10209 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
10210 image analysis. If no field information is available, then this works just
10211 like @samp{U}. This is the default mode.
10214 Both capture and transfer unknown or varying.
10215 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
10219 @section pixdesctest
10221 Pixel format descriptor test filter, mainly useful for internal
10222 testing. The output video should be equal to the input video.
10226 format=monow, pixdesctest
10229 can be used to test the monowhite pixel format descriptor definition.
10233 Enable the specified chain of postprocessing subfilters using libpostproc. This
10234 library should be automatically selected with a GPL build (@code{--enable-gpl}).
10235 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
10236 Each subfilter and some options have a short and a long name that can be used
10237 interchangeably, i.e. dr/dering are the same.
10239 The filters accept the following options:
10243 Set postprocessing subfilters string.
10246 All subfilters share common options to determine their scope:
10250 Honor the quality commands for this subfilter.
10253 Do chrominance filtering, too (default).
10256 Do luminance filtering only (no chrominance).
10259 Do chrominance filtering only (no luminance).
10262 These options can be appended after the subfilter name, separated by a '|'.
10264 Available subfilters are:
10267 @item hb/hdeblock[|difference[|flatness]]
10268 Horizontal deblocking filter
10271 Difference factor where higher values mean more deblocking (default: @code{32}).
10273 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10276 @item vb/vdeblock[|difference[|flatness]]
10277 Vertical deblocking filter
10280 Difference factor where higher values mean more deblocking (default: @code{32}).
10282 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10285 @item ha/hadeblock[|difference[|flatness]]
10286 Accurate horizontal deblocking filter
10289 Difference factor where higher values mean more deblocking (default: @code{32}).
10291 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10294 @item va/vadeblock[|difference[|flatness]]
10295 Accurate vertical deblocking filter
10298 Difference factor where higher values mean more deblocking (default: @code{32}).
10300 Flatness threshold where lower values mean more deblocking (default: @code{39}).
10304 The horizontal and vertical deblocking filters share the difference and
10305 flatness values so you cannot set different horizontal and vertical
10309 @item h1/x1hdeblock
10310 Experimental horizontal deblocking filter
10312 @item v1/x1vdeblock
10313 Experimental vertical deblocking filter
10318 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
10321 larger -> stronger filtering
10323 larger -> stronger filtering
10325 larger -> stronger filtering
10328 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
10331 Stretch luminance to @code{0-255}.
10334 @item lb/linblenddeint
10335 Linear blend deinterlacing filter that deinterlaces the given block by
10336 filtering all lines with a @code{(1 2 1)} filter.
10338 @item li/linipoldeint
10339 Linear interpolating deinterlacing filter that deinterlaces the given block by
10340 linearly interpolating every second line.
10342 @item ci/cubicipoldeint
10343 Cubic interpolating deinterlacing filter deinterlaces the given block by
10344 cubically interpolating every second line.
10346 @item md/mediandeint
10347 Median deinterlacing filter that deinterlaces the given block by applying a
10348 median filter to every second line.
10350 @item fd/ffmpegdeint
10351 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
10352 second line with a @code{(-1 4 2 4 -1)} filter.
10355 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
10356 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
10358 @item fq/forceQuant[|quantizer]
10359 Overrides the quantizer table from the input with the constant quantizer you
10367 Default pp filter combination (@code{hb|a,vb|a,dr|a})
10370 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
10373 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
10376 @subsection Examples
10380 Apply horizontal and vertical deblocking, deringing and automatic
10381 brightness/contrast:
10387 Apply default filters without brightness/contrast correction:
10393 Apply default filters and temporal denoiser:
10395 pp=default/tmpnoise|1|2|3
10399 Apply deblocking on luminance only, and switch vertical deblocking on or off
10400 automatically depending on available CPU time:
10407 Apply Postprocessing filter 7. It is variant of the @ref{spp} filter,
10408 similar to spp = 6 with 7 point DCT, where only the center sample is
10411 The filter accepts the following options:
10415 Force a constant quantization parameter. It accepts an integer in range
10416 0 to 63. If not set, the filter will use the QP from the video stream
10420 Set thresholding mode. Available modes are:
10424 Set hard thresholding.
10426 Set soft thresholding (better de-ringing effect, but likely blurrier).
10428 Set medium thresholding (good results, default).
10434 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
10435 Ratio) between two input videos.
10437 This filter takes in input two input videos, the first input is
10438 considered the "main" source and is passed unchanged to the
10439 output. The second input is used as a "reference" video for computing
10442 Both video inputs must have the same resolution and pixel format for
10443 this filter to work correctly. Also it assumes that both inputs
10444 have the same number of frames, which are compared one by one.
10446 The obtained average PSNR is printed through the logging system.
10448 The filter stores the accumulated MSE (mean squared error) of each
10449 frame, and at the end of the processing it is averaged across all frames
10450 equally, and the following formula is applied to obtain the PSNR:
10453 PSNR = 10*log10(MAX^2/MSE)
10456 Where MAX is the average of the maximum values of each component of the
10459 The description of the accepted parameters follows.
10462 @item stats_file, f
10463 If specified the filter will use the named file to save the PSNR of
10464 each individual frame. When filename equals "-" the data is sent to
10468 The file printed if @var{stats_file} is selected, contains a sequence of
10469 key/value pairs of the form @var{key}:@var{value} for each compared
10472 A description of each shown parameter follows:
10476 sequential number of the input frame, starting from 1
10479 Mean Square Error pixel-by-pixel average difference of the compared
10480 frames, averaged over all the image components.
10482 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
10483 Mean Square Error pixel-by-pixel average difference of the compared
10484 frames for the component specified by the suffix.
10486 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
10487 Peak Signal to Noise ratio of the compared frames for the component
10488 specified by the suffix.
10493 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
10494 [main][ref] psnr="stats_file=stats.log" [out]
10497 On this example the input file being processed is compared with the
10498 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
10499 is stored in @file{stats.log}.
10504 Pulldown reversal (inverse telecine) filter, capable of handling mixed
10505 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
10508 The pullup filter is designed to take advantage of future context in making
10509 its decisions. This filter is stateless in the sense that it does not lock
10510 onto a pattern to follow, but it instead looks forward to the following
10511 fields in order to identify matches and rebuild progressive frames.
10513 To produce content with an even framerate, insert the fps filter after
10514 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
10515 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
10517 The filter accepts the following options:
10524 These options set the amount of "junk" to ignore at the left, right, top, and
10525 bottom of the image, respectively. Left and right are in units of 8 pixels,
10526 while top and bottom are in units of 2 lines.
10527 The default is 8 pixels on each side.
10530 Set the strict breaks. Setting this option to 1 will reduce the chances of
10531 filter generating an occasional mismatched frame, but it may also cause an
10532 excessive number of frames to be dropped during high motion sequences.
10533 Conversely, setting it to -1 will make filter match fields more easily.
10534 This may help processing of video where there is slight blurring between
10535 the fields, but may also cause there to be interlaced frames in the output.
10536 Default value is @code{0}.
10539 Set the metric plane to use. It accepts the following values:
10545 Use chroma blue plane.
10548 Use chroma red plane.
10551 This option may be set to use chroma plane instead of the default luma plane
10552 for doing filter's computations. This may improve accuracy on very clean
10553 source material, but more likely will decrease accuracy, especially if there
10554 is chroma noise (rainbow effect) or any grayscale video.
10555 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
10556 load and make pullup usable in realtime on slow machines.
10559 For best results (without duplicated frames in the output file) it is
10560 necessary to change the output frame rate. For example, to inverse
10561 telecine NTSC input:
10563 ffmpeg -i input -vf pullup -r 24000/1001 ...
10568 Change video quantization parameters (QP).
10570 The filter accepts the following option:
10574 Set expression for quantization parameter.
10577 The expression is evaluated through the eval API and can contain, among others,
10578 the following constants:
10582 1 if index is not 129, 0 otherwise.
10585 Sequentional index starting from -129 to 128.
10588 @subsection Examples
10592 Some equation like:
10600 Flush video frames from internal cache of frames into a random order.
10601 No frame is discarded.
10602 Inspired by @ref{frei0r} nervous filter.
10606 Set size in number of frames of internal cache, in range from @code{2} to
10607 @code{512}. Default is @code{30}.
10610 Set seed for random number generator, must be an integer included between
10611 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
10612 less than @code{0}, the filter will try to use a good random seed on a
10618 Remap pixels using 2nd: Xmap and 3rd: Ymap input video stream.
10620 Destination pixel at position (X, Y) will be picked from source (x, y) position
10621 where x = Xmap(X, Y) and y = Ymap(X, Y). If mapping values are out of range, zero
10622 value for pixel will be used for destination pixel.
10624 Xmap and Ymap input video streams must be of same dimensions. Output video stream
10625 will have Xmap/Ymap video stream dimensions.
10626 Xmap and Ymap input video streams are 16bit depth, single channel.
10628 @section removegrain
10630 The removegrain filter is a spatial denoiser for progressive video.
10634 Set mode for the first plane.
10637 Set mode for the second plane.
10640 Set mode for the third plane.
10643 Set mode for the fourth plane.
10646 Range of mode is from 0 to 24. Description of each mode follows:
10650 Leave input plane unchanged. Default.
10653 Clips the pixel with the minimum and maximum of the 8 neighbour pixels.
10656 Clips the pixel with the second minimum and maximum of the 8 neighbour pixels.
10659 Clips the pixel with the third minimum and maximum of the 8 neighbour pixels.
10662 Clips the pixel with the fourth minimum and maximum of the 8 neighbour pixels.
10663 This is equivalent to a median filter.
10666 Line-sensitive clipping giving the minimal change.
10669 Line-sensitive clipping, intermediate.
10672 Line-sensitive clipping, intermediate.
10675 Line-sensitive clipping, intermediate.
10678 Line-sensitive clipping on a line where the neighbours pixels are the closest.
10681 Replaces the target pixel with the closest neighbour.
10684 [1 2 1] horizontal and vertical kernel blur.
10690 Bob mode, interpolates top field from the line where the neighbours
10691 pixels are the closest.
10694 Bob mode, interpolates bottom field from the line where the neighbours
10695 pixels are the closest.
10698 Bob mode, interpolates top field. Same as 13 but with a more complicated
10699 interpolation formula.
10702 Bob mode, interpolates bottom field. Same as 14 but with a more complicated
10703 interpolation formula.
10706 Clips the pixel with the minimum and maximum of respectively the maximum and
10707 minimum of each pair of opposite neighbour pixels.
10710 Line-sensitive clipping using opposite neighbours whose greatest distance from
10711 the current pixel is minimal.
10714 Replaces the pixel with the average of its 8 neighbours.
10717 Averages the 9 pixels ([1 1 1] horizontal and vertical blur).
10720 Clips pixels using the averages of opposite neighbour.
10723 Same as mode 21 but simpler and faster.
10726 Small edge and halo removal, but reputed useless.
10732 @section removelogo
10734 Suppress a TV station logo, using an image file to determine which
10735 pixels comprise the logo. It works by filling in the pixels that
10736 comprise the logo with neighboring pixels.
10738 The filter accepts the following options:
10742 Set the filter bitmap file, which can be any image format supported by
10743 libavformat. The width and height of the image file must match those of the
10744 video stream being processed.
10747 Pixels in the provided bitmap image with a value of zero are not
10748 considered part of the logo, non-zero pixels are considered part of
10749 the logo. If you use white (255) for the logo and black (0) for the
10750 rest, you will be safe. For making the filter bitmap, it is
10751 recommended to take a screen capture of a black frame with the logo
10752 visible, and then using a threshold filter followed by the erode
10753 filter once or twice.
10755 If needed, little splotches can be fixed manually. Remember that if
10756 logo pixels are not covered, the filter quality will be much
10757 reduced. Marking too many pixels as part of the logo does not hurt as
10758 much, but it will increase the amount of blurring needed to cover over
10759 the image and will destroy more information than necessary, and extra
10760 pixels will slow things down on a large logo.
10762 @section repeatfields
10764 This filter uses the repeat_field flag from the Video ES headers and hard repeats
10765 fields based on its value.
10767 @section reverse, areverse
10771 Warning: This filter requires memory to buffer the entire clip, so trimming
10774 @subsection Examples
10778 Take the first 5 seconds of a clip, and reverse it.
10786 Rotate video by an arbitrary angle expressed in radians.
10788 The filter accepts the following options:
10790 A description of the optional parameters follows.
10793 Set an expression for the angle by which to rotate the input video
10794 clockwise, expressed as a number of radians. A negative value will
10795 result in a counter-clockwise rotation. By default it is set to "0".
10797 This expression is evaluated for each frame.
10800 Set the output width expression, default value is "iw".
10801 This expression is evaluated just once during configuration.
10804 Set the output height expression, default value is "ih".
10805 This expression is evaluated just once during configuration.
10808 Enable bilinear interpolation if set to 1, a value of 0 disables
10809 it. Default value is 1.
10812 Set the color used to fill the output area not covered by the rotated
10813 image. For the general syntax of this option, check the "Color" section in the
10814 ffmpeg-utils manual. If the special value "none" is selected then no
10815 background is printed (useful for example if the background is never shown).
10817 Default value is "black".
10820 The expressions for the angle and the output size can contain the
10821 following constants and functions:
10825 sequential number of the input frame, starting from 0. It is always NAN
10826 before the first frame is filtered.
10829 time in seconds of the input frame, it is set to 0 when the filter is
10830 configured. It is always NAN before the first frame is filtered.
10834 horizontal and vertical chroma subsample values. For example for the
10835 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
10839 the input video width and height
10843 the output width and height, that is the size of the padded area as
10844 specified by the @var{width} and @var{height} expressions
10848 the minimal width/height required for completely containing the input
10849 video rotated by @var{a} radians.
10851 These are only available when computing the @option{out_w} and
10852 @option{out_h} expressions.
10855 @subsection Examples
10859 Rotate the input by PI/6 radians clockwise:
10865 Rotate the input by PI/6 radians counter-clockwise:
10871 Rotate the input by 45 degrees clockwise:
10877 Apply a constant rotation with period T, starting from an angle of PI/3:
10879 rotate=PI/3+2*PI*t/T
10883 Make the input video rotation oscillating with a period of T
10884 seconds and an amplitude of A radians:
10886 rotate=A*sin(2*PI/T*t)
10890 Rotate the video, output size is chosen so that the whole rotating
10891 input video is always completely contained in the output:
10893 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
10897 Rotate the video, reduce the output size so that no background is ever
10900 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
10904 @subsection Commands
10906 The filter supports the following commands:
10910 Set the angle expression.
10911 The command accepts the same syntax of the corresponding option.
10913 If the specified expression is not valid, it is kept at its current
10919 Apply Shape Adaptive Blur.
10921 The filter accepts the following options:
10924 @item luma_radius, lr
10925 Set luma blur filter strength, must be a value in range 0.1-4.0, default
10926 value is 1.0. A greater value will result in a more blurred image, and
10927 in slower processing.
10929 @item luma_pre_filter_radius, lpfr
10930 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
10933 @item luma_strength, ls
10934 Set luma maximum difference between pixels to still be considered, must
10935 be a value in the 0.1-100.0 range, default value is 1.0.
10937 @item chroma_radius, cr
10938 Set chroma blur filter strength, must be a value in range 0.1-4.0. A
10939 greater value will result in a more blurred image, and in slower
10942 @item chroma_pre_filter_radius, cpfr
10943 Set chroma pre-filter radius, must be a value in the 0.1-2.0 range.
10945 @item chroma_strength, cs
10946 Set chroma maximum difference between pixels to still be considered,
10947 must be a value in the 0.1-100.0 range.
10950 Each chroma option value, if not explicitly specified, is set to the
10951 corresponding luma option value.
10956 Scale (resize) the input video, using the libswscale library.
10958 The scale filter forces the output display aspect ratio to be the same
10959 of the input, by changing the output sample aspect ratio.
10961 If the input image format is different from the format requested by
10962 the next filter, the scale filter will convert the input to the
10965 @subsection Options
10966 The filter accepts the following options, or any of the options
10967 supported by the libswscale scaler.
10969 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
10970 the complete list of scaler options.
10975 Set the output video dimension expression. Default value is the input
10978 If the value is 0, the input width is used for the output.
10980 If one of the values is -1, the scale filter will use a value that
10981 maintains the aspect ratio of the input image, calculated from the
10982 other specified dimension. If both of them are -1, the input size is
10985 If one of the values is -n with n > 1, the scale filter will also use a value
10986 that maintains the aspect ratio of the input image, calculated from the other
10987 specified dimension. After that it will, however, make sure that the calculated
10988 dimension is divisible by n and adjust the value if necessary.
10990 See below for the list of accepted constants for use in the dimension
10994 Specify when to evaluate @var{width} and @var{height} expression. It accepts the following values:
10998 Only evaluate expressions once during the filter initialization or when a command is processed.
11001 Evaluate expressions for each incoming frame.
11005 Default value is @samp{init}.
11009 Set the interlacing mode. It accepts the following values:
11013 Force interlaced aware scaling.
11016 Do not apply interlaced scaling.
11019 Select interlaced aware scaling depending on whether the source frames
11020 are flagged as interlaced or not.
11023 Default value is @samp{0}.
11026 Set libswscale scaling flags. See
11027 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11028 complete list of values. If not explicitly specified the filter applies
11032 @item param0, param1
11033 Set libswscale input parameters for scaling algorithms that need them. See
11034 @ref{sws_params,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
11035 complete documentation. If not explicitly specified the filter applies
11041 Set the video size. For the syntax of this option, check the
11042 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11044 @item in_color_matrix
11045 @item out_color_matrix
11046 Set in/output YCbCr color space type.
11048 This allows the autodetected value to be overridden as well as allows forcing
11049 a specific value used for the output and encoder.
11051 If not specified, the color space type depends on the pixel format.
11057 Choose automatically.
11060 Format conforming to International Telecommunication Union (ITU)
11061 Recommendation BT.709.
11064 Set color space conforming to the United States Federal Communications
11065 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
11068 Set color space conforming to:
11072 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
11075 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
11078 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
11083 Set color space conforming to SMPTE ST 240:1999.
11088 Set in/output YCbCr sample range.
11090 This allows the autodetected value to be overridden as well as allows forcing
11091 a specific value used for the output and encoder. If not specified, the
11092 range depends on the pixel format. Possible values:
11096 Choose automatically.
11099 Set full range (0-255 in case of 8-bit luma).
11102 Set "MPEG" range (16-235 in case of 8-bit luma).
11105 @item force_original_aspect_ratio
11106 Enable decreasing or increasing output video width or height if necessary to
11107 keep the original aspect ratio. Possible values:
11111 Scale the video as specified and disable this feature.
11114 The output video dimensions will automatically be decreased if needed.
11117 The output video dimensions will automatically be increased if needed.
11121 One useful instance of this option is that when you know a specific device's
11122 maximum allowed resolution, you can use this to limit the output video to
11123 that, while retaining the aspect ratio. For example, device A allows
11124 1280x720 playback, and your video is 1920x800. Using this option (set it to
11125 decrease) and specifying 1280x720 to the command line makes the output
11128 Please note that this is a different thing than specifying -1 for @option{w}
11129 or @option{h}, you still need to specify the output resolution for this option
11134 The values of the @option{w} and @option{h} options are expressions
11135 containing the following constants:
11140 The input width and height
11144 These are the same as @var{in_w} and @var{in_h}.
11148 The output (scaled) width and height
11152 These are the same as @var{out_w} and @var{out_h}
11155 The same as @var{iw} / @var{ih}
11158 input sample aspect ratio
11161 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
11165 horizontal and vertical input chroma subsample values. For example for the
11166 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11170 horizontal and vertical output chroma subsample values. For example for the
11171 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11174 @subsection Examples
11178 Scale the input video to a size of 200x100
11183 This is equivalent to:
11194 Specify a size abbreviation for the output size:
11199 which can also be written as:
11205 Scale the input to 2x:
11207 scale=w=2*iw:h=2*ih
11211 The above is the same as:
11213 scale=2*in_w:2*in_h
11217 Scale the input to 2x with forced interlaced scaling:
11219 scale=2*iw:2*ih:interl=1
11223 Scale the input to half size:
11225 scale=w=iw/2:h=ih/2
11229 Increase the width, and set the height to the same size:
11235 Seek Greek harmony:
11242 Increase the height, and set the width to 3/2 of the height:
11244 scale=w=3/2*oh:h=3/5*ih
11248 Increase the size, making the size a multiple of the chroma
11251 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
11255 Increase the width to a maximum of 500 pixels,
11256 keeping the same aspect ratio as the input:
11258 scale=w='min(500\, iw*3/2):h=-1'
11262 @subsection Commands
11264 This filter supports the following commands:
11268 Set the output video dimension expression.
11269 The command accepts the same syntax of the corresponding option.
11271 If the specified expression is not valid, it is kept at its current
11277 Scale (resize) the input video, based on a reference video.
11279 See the scale filter for available options, scale2ref supports the same but
11280 uses the reference video instead of the main input as basis.
11282 @subsection Examples
11286 Scale a subtitle stream to match the main video in size before overlaying
11288 'scale2ref[b][a];[a][b]overlay'
11292 @anchor{selectivecolor}
11293 @section selectivecolor
11295 Adjust cyan, magenta, yellow and black (CMYK) to certain ranges of colors (such
11296 as "reds", "yellows", "greens", "cyans", ...). The adjustment range is defined
11297 by the "purity" of the color (that is, how saturated it already is).
11299 This filter is similar to the Adobe Photoshop Selective Color tool.
11301 The filter accepts the following options:
11304 @item correction_method
11305 Select color correction method.
11307 Available values are:
11310 Specified adjustments are applied "as-is" (added/subtracted to original pixel
11313 Specified adjustments are relative to the original component value.
11315 Default is @code{absolute}.
11317 Adjustments for red pixels (pixels where the red component is the maximum)
11319 Adjustments for yellow pixels (pixels where the blue component is the minimum)
11321 Adjustments for green pixels (pixels where the green component is the maximum)
11323 Adjustments for cyan pixels (pixels where the red component is the minimum)
11325 Adjustments for blue pixels (pixels where the blue component is the maximum)
11327 Adjustments for magenta pixels (pixels where the green component is the minimum)
11329 Adjustments for white pixels (pixels where all components are greater than 128)
11331 Adjustments for all pixels except pure black and pure white
11333 Adjustments for black pixels (pixels where all components are lesser than 128)
11335 Specify a Photoshop selective color file (@code{.asv}) to import the settings from.
11338 All the adjustment settings (@option{reds}, @option{yellows}, ...) accept up to
11339 4 space separated floating point adjustment values in the [-1,1] range,
11340 respectively to adjust the amount of cyan, magenta, yellow and black for the
11341 pixels of its range.
11343 @subsection Examples
11347 Increase cyan by 50% and reduce yellow by 33% in every green areas, and
11348 increase magenta by 27% in blue areas:
11350 selectivecolor=greens=.5 0 -.33 0:blues=0 .27
11354 Use a Photoshop selective color preset:
11356 selectivecolor=psfile=MySelectiveColorPresets/Misty.asv
11360 @section separatefields
11362 The @code{separatefields} takes a frame-based video input and splits
11363 each frame into its components fields, producing a new half height clip
11364 with twice the frame rate and twice the frame count.
11366 This filter use field-dominance information in frame to decide which
11367 of each pair of fields to place first in the output.
11368 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
11370 @section setdar, setsar
11372 The @code{setdar} filter sets the Display Aspect Ratio for the filter
11375 This is done by changing the specified Sample (aka Pixel) Aspect
11376 Ratio, according to the following equation:
11378 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
11381 Keep in mind that the @code{setdar} filter does not modify the pixel
11382 dimensions of the video frame. Also, the display aspect ratio set by
11383 this filter may be changed by later filters in the filterchain,
11384 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
11387 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
11388 the filter output video.
11390 Note that as a consequence of the application of this filter, the
11391 output display aspect ratio will change according to the equation
11394 Keep in mind that the sample aspect ratio set by the @code{setsar}
11395 filter may be changed by later filters in the filterchain, e.g. if
11396 another "setsar" or a "setdar" filter is applied.
11398 It accepts the following parameters:
11401 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
11402 Set the aspect ratio used by the filter.
11404 The parameter can be a floating point number string, an expression, or
11405 a string of the form @var{num}:@var{den}, where @var{num} and
11406 @var{den} are the numerator and denominator of the aspect ratio. If
11407 the parameter is not specified, it is assumed the value "0".
11408 In case the form "@var{num}:@var{den}" is used, the @code{:} character
11412 Set the maximum integer value to use for expressing numerator and
11413 denominator when reducing the expressed aspect ratio to a rational.
11414 Default value is @code{100}.
11418 The parameter @var{sar} is an expression containing
11419 the following constants:
11423 These are approximated values for the mathematical constants e
11424 (Euler's number), pi (Greek pi), and phi (the golden ratio).
11427 The input width and height.
11430 These are the same as @var{w} / @var{h}.
11433 The input sample aspect ratio.
11436 The input display aspect ratio. It is the same as
11437 (@var{w} / @var{h}) * @var{sar}.
11440 Horizontal and vertical chroma subsample values. For example, for the
11441 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
11444 @subsection Examples
11449 To change the display aspect ratio to 16:9, specify one of the following:
11457 To change the sample aspect ratio to 10:11, specify:
11463 To set a display aspect ratio of 16:9, and specify a maximum integer value of
11464 1000 in the aspect ratio reduction, use the command:
11466 setdar=ratio=16/9:max=1000
11474 Force field for the output video frame.
11476 The @code{setfield} filter marks the interlace type field for the
11477 output frames. It does not change the input frame, but only sets the
11478 corresponding property, which affects how the frame is treated by
11479 following filters (e.g. @code{fieldorder} or @code{yadif}).
11481 The filter accepts the following options:
11486 Available values are:
11490 Keep the same field property.
11493 Mark the frame as bottom-field-first.
11496 Mark the frame as top-field-first.
11499 Mark the frame as progressive.
11505 Show a line containing various information for each input video frame.
11506 The input video is not modified.
11508 The shown line contains a sequence of key/value pairs of the form
11509 @var{key}:@var{value}.
11511 The following values are shown in the output:
11515 The (sequential) number of the input frame, starting from 0.
11518 The Presentation TimeStamp of the input frame, expressed as a number of
11519 time base units. The time base unit depends on the filter input pad.
11522 The Presentation TimeStamp of the input frame, expressed as a number of
11526 The position of the frame in the input stream, or -1 if this information is
11527 unavailable and/or meaningless (for example in case of synthetic video).
11530 The pixel format name.
11533 The sample aspect ratio of the input frame, expressed in the form
11534 @var{num}/@var{den}.
11537 The size of the input frame. For the syntax of this option, check the
11538 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
11541 The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
11542 for bottom field first).
11545 This is 1 if the frame is a key frame, 0 otherwise.
11548 The picture type of the input frame ("I" for an I-frame, "P" for a
11549 P-frame, "B" for a B-frame, or "?" for an unknown type).
11550 Also refer to the documentation of the @code{AVPictureType} enum and of
11551 the @code{av_get_picture_type_char} function defined in
11552 @file{libavutil/avutil.h}.
11555 The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
11557 @item plane_checksum
11558 The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
11559 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
11562 @section showpalette
11564 Displays the 256 colors palette of each frame. This filter is only relevant for
11565 @var{pal8} pixel format frames.
11567 It accepts the following option:
11571 Set the size of the box used to represent one palette color entry. Default is
11572 @code{30} (for a @code{30x30} pixel box).
11575 @section shuffleframes
11577 Reorder and/or duplicate video frames.
11579 It accepts the following parameters:
11583 Set the destination indexes of input frames.
11584 This is space or '|' separated list of indexes that maps input frames to output
11585 frames. Number of indexes also sets maximal value that each index may have.
11588 The first frame has the index 0. The default is to keep the input unchanged.
11590 Swap second and third frame of every three frames of the input:
11592 ffmpeg -i INPUT -vf "shuffleframes=0 2 1" OUTPUT
11595 @section shuffleplanes
11597 Reorder and/or duplicate video planes.
11599 It accepts the following parameters:
11604 The index of the input plane to be used as the first output plane.
11607 The index of the input plane to be used as the second output plane.
11610 The index of the input plane to be used as the third output plane.
11613 The index of the input plane to be used as the fourth output plane.
11617 The first plane has the index 0. The default is to keep the input unchanged.
11619 Swap the second and third planes of the input:
11621 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
11624 @anchor{signalstats}
11625 @section signalstats
11626 Evaluate various visual metrics that assist in determining issues associated
11627 with the digitization of analog video media.
11629 By default the filter will log these metadata values:
11633 Display the minimal Y value contained within the input frame. Expressed in
11637 Display the Y value at the 10% percentile within the input frame. Expressed in
11641 Display the average Y value within the input frame. Expressed in range of
11645 Display the Y value at the 90% percentile within the input frame. Expressed in
11649 Display the maximum Y value contained within the input frame. Expressed in
11653 Display the minimal U value contained within the input frame. Expressed in
11657 Display the U value at the 10% percentile within the input frame. Expressed in
11661 Display the average U value within the input frame. Expressed in range of
11665 Display the U value at the 90% percentile within the input frame. Expressed in
11669 Display the maximum U value contained within the input frame. Expressed in
11673 Display the minimal V value contained within the input frame. Expressed in
11677 Display the V value at the 10% percentile within the input frame. Expressed in
11681 Display the average V value within the input frame. Expressed in range of
11685 Display the V value at the 90% percentile within the input frame. Expressed in
11689 Display the maximum V value contained within the input frame. Expressed in
11693 Display the minimal saturation value contained within the input frame.
11694 Expressed in range of [0-~181.02].
11697 Display the saturation value at the 10% percentile within the input frame.
11698 Expressed in range of [0-~181.02].
11701 Display the average saturation value within the input frame. Expressed in range
11705 Display the saturation value at the 90% percentile within the input frame.
11706 Expressed in range of [0-~181.02].
11709 Display the maximum saturation value contained within the input frame.
11710 Expressed in range of [0-~181.02].
11713 Display the median value for hue within the input frame. Expressed in range of
11717 Display the average value for hue within the input frame. Expressed in range of
11721 Display the average of sample value difference between all values of the Y
11722 plane in the current frame and corresponding values of the previous input frame.
11723 Expressed in range of [0-255].
11726 Display the average of sample value difference between all values of the U
11727 plane in the current frame and corresponding values of the previous input frame.
11728 Expressed in range of [0-255].
11731 Display the average of sample value difference between all values of the V
11732 plane in the current frame and corresponding values of the previous input frame.
11733 Expressed in range of [0-255].
11736 The filter accepts the following options:
11742 @option{stat} specify an additional form of image analysis.
11743 @option{out} output video with the specified type of pixel highlighted.
11745 Both options accept the following values:
11749 Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
11750 unlike the neighboring pixels of the same field. Examples of temporal outliers
11751 include the results of video dropouts, head clogs, or tape tracking issues.
11754 Identify @var{vertical line repetition}. Vertical line repetition includes
11755 similar rows of pixels within a frame. In born-digital video vertical line
11756 repetition is common, but this pattern is uncommon in video digitized from an
11757 analog source. When it occurs in video that results from the digitization of an
11758 analog source it can indicate concealment from a dropout compensator.
11761 Identify pixels that fall outside of legal broadcast range.
11765 Set the highlight color for the @option{out} option. The default color is
11769 @subsection Examples
11773 Output data of various video metrics:
11775 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
11779 Output specific data about the minimum and maximum values of the Y plane per frame:
11781 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
11785 Playback video while highlighting pixels that are outside of broadcast range in red.
11787 ffplay example.mov -vf signalstats="out=brng:color=red"
11791 Playback video with signalstats metadata drawn over the frame.
11793 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
11796 The contents of signalstat_drawtext.txt used in the command are:
11799 Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
11800 U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
11801 V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
11802 saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
11810 Blur the input video without impacting the outlines.
11812 It accepts the following options:
11815 @item luma_radius, lr
11816 Set the luma radius. The option value must be a float number in
11817 the range [0.1,5.0] that specifies the variance of the gaussian filter
11818 used to blur the image (slower if larger). Default value is 1.0.
11820 @item luma_strength, ls
11821 Set the luma strength. The option value must be a float number
11822 in the range [-1.0,1.0] that configures the blurring. A value included
11823 in [0.0,1.0] will blur the image whereas a value included in
11824 [-1.0,0.0] will sharpen the image. Default value is 1.0.
11826 @item luma_threshold, lt
11827 Set the luma threshold used as a coefficient to determine
11828 whether a pixel should be blurred or not. The option value must be an
11829 integer in the range [-30,30]. A value of 0 will filter all the image,
11830 a value included in [0,30] will filter flat areas and a value included
11831 in [-30,0] will filter edges. Default value is 0.
11833 @item chroma_radius, cr
11834 Set the chroma radius. The option value must be a float number in
11835 the range [0.1,5.0] that specifies the variance of the gaussian filter
11836 used to blur the image (slower if larger). Default value is 1.0.
11838 @item chroma_strength, cs
11839 Set the chroma strength. The option value must be a float number
11840 in the range [-1.0,1.0] that configures the blurring. A value included
11841 in [0.0,1.0] will blur the image whereas a value included in
11842 [-1.0,0.0] will sharpen the image. Default value is 1.0.
11844 @item chroma_threshold, ct
11845 Set the chroma threshold used as a coefficient to determine
11846 whether a pixel should be blurred or not. The option value must be an
11847 integer in the range [-30,30]. A value of 0 will filter all the image,
11848 a value included in [0,30] will filter flat areas and a value included
11849 in [-30,0] will filter edges. Default value is 0.
11852 If a chroma option is not explicitly set, the corresponding luma value
11857 Obtain the SSIM (Structural SImilarity Metric) between two input videos.
11859 This filter takes in input two input videos, the first input is
11860 considered the "main" source and is passed unchanged to the
11861 output. The second input is used as a "reference" video for computing
11864 Both video inputs must have the same resolution and pixel format for
11865 this filter to work correctly. Also it assumes that both inputs
11866 have the same number of frames, which are compared one by one.
11868 The filter stores the calculated SSIM of each frame.
11870 The description of the accepted parameters follows.
11873 @item stats_file, f
11874 If specified the filter will use the named file to save the SSIM of
11875 each individual frame. When filename equals "-" the data is sent to
11879 The file printed if @var{stats_file} is selected, contains a sequence of
11880 key/value pairs of the form @var{key}:@var{value} for each compared
11883 A description of each shown parameter follows:
11887 sequential number of the input frame, starting from 1
11889 @item Y, U, V, R, G, B
11890 SSIM of the compared frames for the component specified by the suffix.
11893 SSIM of the compared frames for the whole frame.
11896 Same as above but in dB representation.
11901 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
11902 [main][ref] ssim="stats_file=stats.log" [out]
11905 On this example the input file being processed is compared with the
11906 reference file @file{ref_movie.mpg}. The SSIM of each individual frame
11907 is stored in @file{stats.log}.
11909 Another example with both psnr and ssim at same time:
11911 ffmpeg -i main.mpg -i ref.mpg -lavfi "ssim;[0:v][1:v]psnr" -f null -
11916 Convert between different stereoscopic image formats.
11918 The filters accept the following options:
11922 Set stereoscopic image format of input.
11924 Available values for input image formats are:
11927 side by side parallel (left eye left, right eye right)
11930 side by side crosseye (right eye left, left eye right)
11933 side by side parallel with half width resolution
11934 (left eye left, right eye right)
11937 side by side crosseye with half width resolution
11938 (right eye left, left eye right)
11941 above-below (left eye above, right eye below)
11944 above-below (right eye above, left eye below)
11947 above-below with half height resolution
11948 (left eye above, right eye below)
11951 above-below with half height resolution
11952 (right eye above, left eye below)
11955 alternating frames (left eye first, right eye second)
11958 alternating frames (right eye first, left eye second)
11961 interleaved rows (left eye has top row, right eye starts on next row)
11964 interleaved rows (right eye has top row, left eye starts on next row)
11967 interleaved columns, left eye first
11970 interleaved columns, right eye first
11972 Default value is @samp{sbsl}.
11976 Set stereoscopic image format of output.
11980 side by side parallel (left eye left, right eye right)
11983 side by side crosseye (right eye left, left eye right)
11986 side by side parallel with half width resolution
11987 (left eye left, right eye right)
11990 side by side crosseye with half width resolution
11991 (right eye left, left eye right)
11994 above-below (left eye above, right eye below)
11997 above-below (right eye above, left eye below)
12000 above-below with half height resolution
12001 (left eye above, right eye below)
12004 above-below with half height resolution
12005 (right eye above, left eye below)
12008 alternating frames (left eye first, right eye second)
12011 alternating frames (right eye first, left eye second)
12014 interleaved rows (left eye has top row, right eye starts on next row)
12017 interleaved rows (right eye has top row, left eye starts on next row)
12020 anaglyph red/blue gray
12021 (red filter on left eye, blue filter on right eye)
12024 anaglyph red/green gray
12025 (red filter on left eye, green filter on right eye)
12028 anaglyph red/cyan gray
12029 (red filter on left eye, cyan filter on right eye)
12032 anaglyph red/cyan half colored
12033 (red filter on left eye, cyan filter on right eye)
12036 anaglyph red/cyan color
12037 (red filter on left eye, cyan filter on right eye)
12040 anaglyph red/cyan color optimized with the least squares projection of dubois
12041 (red filter on left eye, cyan filter on right eye)
12044 anaglyph green/magenta gray
12045 (green filter on left eye, magenta filter on right eye)
12048 anaglyph green/magenta half colored
12049 (green filter on left eye, magenta filter on right eye)
12052 anaglyph green/magenta colored
12053 (green filter on left eye, magenta filter on right eye)
12056 anaglyph green/magenta color optimized with the least squares projection of dubois
12057 (green filter on left eye, magenta filter on right eye)
12060 anaglyph yellow/blue gray
12061 (yellow filter on left eye, blue filter on right eye)
12064 anaglyph yellow/blue half colored
12065 (yellow filter on left eye, blue filter on right eye)
12068 anaglyph yellow/blue colored
12069 (yellow filter on left eye, blue filter on right eye)
12072 anaglyph yellow/blue color optimized with the least squares projection of dubois
12073 (yellow filter on left eye, blue filter on right eye)
12076 mono output (left eye only)
12079 mono output (right eye only)
12082 checkerboard, left eye first
12085 checkerboard, right eye first
12088 interleaved columns, left eye first
12091 interleaved columns, right eye first
12094 Default value is @samp{arcd}.
12097 @subsection Examples
12101 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
12107 Convert input video from above below (left eye above, right eye below) to side by side crosseye.
12113 @section streamselect, astreamselect
12114 Select video or audio streams.
12116 The filter accepts the following options:
12120 Set number of inputs. Default is 2.
12123 Set input indexes to remap to outputs.
12126 @subsection Commands
12128 The @code{streamselect} and @code{astreamselect} filter supports the following
12133 Set input indexes to remap to outputs.
12136 @subsection Examples
12140 Select first 5 seconds 1st stream and rest of time 2nd stream:
12142 sendcmd='5.0 streamselect map 1',streamselect=inputs=2:map=0
12146 Same as above, but for audio:
12148 asendcmd='5.0 astreamselect map 1',astreamselect=inputs=2:map=0
12155 Apply a simple postprocessing filter that compresses and decompresses the image
12156 at several (or - in the case of @option{quality} level @code{6} - all) shifts
12157 and average the results.
12159 The filter accepts the following options:
12163 Set quality. This option defines the number of levels for averaging. It accepts
12164 an integer in the range 0-6. If set to @code{0}, the filter will have no
12165 effect. A value of @code{6} means the higher quality. For each increment of
12166 that value the speed drops by a factor of approximately 2. Default value is
12170 Force a constant quantization parameter. If not set, the filter will use the QP
12171 from the video stream (if available).
12174 Set thresholding mode. Available modes are:
12178 Set hard thresholding (default).
12180 Set soft thresholding (better de-ringing effect, but likely blurrier).
12183 @item use_bframe_qp
12184 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
12185 option may cause flicker since the B-Frames have often larger QP. Default is
12186 @code{0} (not enabled).
12192 Draw subtitles on top of input video using the libass library.
12194 To enable compilation of this filter you need to configure FFmpeg with
12195 @code{--enable-libass}. This filter also requires a build with libavcodec and
12196 libavformat to convert the passed subtitles file to ASS (Advanced Substation
12197 Alpha) subtitles format.
12199 The filter accepts the following options:
12203 Set the filename of the subtitle file to read. It must be specified.
12205 @item original_size
12206 Specify the size of the original video, the video for which the ASS file
12207 was composed. For the syntax of this option, check the
12208 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12209 Due to a misdesign in ASS aspect ratio arithmetic, this is necessary to
12210 correctly scale the fonts if the aspect ratio has been changed.
12213 Set a directory path containing fonts that can be used by the filter.
12214 These fonts will be used in addition to whatever the font provider uses.
12217 Set subtitles input character encoding. @code{subtitles} filter only. Only
12218 useful if not UTF-8.
12220 @item stream_index, si
12221 Set subtitles stream index. @code{subtitles} filter only.
12224 Override default style or script info parameters of the subtitles. It accepts a
12225 string containing ASS style format @code{KEY=VALUE} couples separated by ",".
12228 If the first key is not specified, it is assumed that the first value
12229 specifies the @option{filename}.
12231 For example, to render the file @file{sub.srt} on top of the input
12232 video, use the command:
12237 which is equivalent to:
12239 subtitles=filename=sub.srt
12242 To render the default subtitles stream from file @file{video.mkv}, use:
12244 subtitles=video.mkv
12247 To render the second subtitles stream from that file, use:
12249 subtitles=video.mkv:si=1
12252 To make the subtitles stream from @file{sub.srt} appear in transparent green
12253 @code{DejaVu Serif}, use:
12255 subtitles=sub.srt:force_style='FontName=DejaVu Serif,PrimaryColour=&HAA00FF00'
12258 @section super2xsai
12260 Scale the input by 2x and smooth using the Super2xSaI (Scale and
12261 Interpolate) pixel art scaling algorithm.
12263 Useful for enlarging pixel art images without reducing sharpness.
12267 Swap two rectangular objects in video.
12269 This filter accepts the following options:
12279 Set 1st rect x coordinate.
12282 Set 1st rect y coordinate.
12285 Set 2nd rect x coordinate.
12288 Set 2nd rect y coordinate.
12290 All expressions are evaluated once for each frame.
12293 The all options are expressions containing the following constants:
12298 The input width and height.
12301 same as @var{w} / @var{h}
12304 input sample aspect ratio
12307 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
12310 The number of the input frame, starting from 0.
12313 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
12316 the position in the file of the input frame, NAN if unknown
12324 Apply telecine process to the video.
12326 This filter accepts the following options:
12335 The default value is @code{top}.
12339 A string of numbers representing the pulldown pattern you wish to apply.
12340 The default value is @code{23}.
12344 Some typical patterns:
12349 24p: 2332 (preferred)
12356 24p: 222222222223 ("Euro pulldown")
12362 Select the most representative frame in a given sequence of consecutive frames.
12364 The filter accepts the following options:
12368 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
12369 will pick one of them, and then handle the next batch of @var{n} frames until
12370 the end. Default is @code{100}.
12373 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
12374 value will result in a higher memory usage, so a high value is not recommended.
12376 @subsection Examples
12380 Extract one picture each 50 frames:
12386 Complete example of a thumbnail creation with @command{ffmpeg}:
12388 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
12394 Tile several successive frames together.
12396 The filter accepts the following options:
12401 Set the grid size (i.e. the number of lines and columns). For the syntax of
12402 this option, check the
12403 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
12406 Set the maximum number of frames to render in the given area. It must be less
12407 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
12408 the area will be used.
12411 Set the outer border margin in pixels.
12414 Set the inner border thickness (i.e. the number of pixels between frames). For
12415 more advanced padding options (such as having different values for the edges),
12416 refer to the pad video filter.
12419 Specify the color of the unused area. For the syntax of this option, check the
12420 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
12424 @subsection Examples
12428 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
12430 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
12432 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
12433 duplicating each output frame to accommodate the originally detected frame
12437 Display @code{5} pictures in an area of @code{3x2} frames,
12438 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
12439 mixed flat and named options:
12441 tile=3x2:nb_frames=5:padding=7:margin=2
12445 @section tinterlace
12447 Perform various types of temporal field interlacing.
12449 Frames are counted starting from 1, so the first input frame is
12452 The filter accepts the following options:
12457 Specify the mode of the interlacing. This option can also be specified
12458 as a value alone. See below for a list of values for this option.
12460 Available values are:
12464 Move odd frames into the upper field, even into the lower field,
12465 generating a double height frame at half frame rate.
12469 Frame 1 Frame 2 Frame 3 Frame 4
12471 11111 22222 33333 44444
12472 11111 22222 33333 44444
12473 11111 22222 33333 44444
12474 11111 22222 33333 44444
12488 Only output even frames, odd frames are dropped, generating a frame with
12489 unchanged height at half frame rate.
12494 Frame 1 Frame 2 Frame 3 Frame 4
12496 11111 22222 33333 44444
12497 11111 22222 33333 44444
12498 11111 22222 33333 44444
12499 11111 22222 33333 44444
12509 Only output odd frames, even frames are dropped, generating a frame with
12510 unchanged height at half frame rate.
12515 Frame 1 Frame 2 Frame 3 Frame 4
12517 11111 22222 33333 44444
12518 11111 22222 33333 44444
12519 11111 22222 33333 44444
12520 11111 22222 33333 44444
12530 Expand each frame to full height, but pad alternate lines with black,
12531 generating a frame with double height at the same input frame rate.
12536 Frame 1 Frame 2 Frame 3 Frame 4
12538 11111 22222 33333 44444
12539 11111 22222 33333 44444
12540 11111 22222 33333 44444
12541 11111 22222 33333 44444
12544 11111 ..... 33333 .....
12545 ..... 22222 ..... 44444
12546 11111 ..... 33333 .....
12547 ..... 22222 ..... 44444
12548 11111 ..... 33333 .....
12549 ..... 22222 ..... 44444
12550 11111 ..... 33333 .....
12551 ..... 22222 ..... 44444
12555 @item interleave_top, 4
12556 Interleave the upper field from odd frames with the lower field from
12557 even frames, generating a frame with unchanged height at half frame rate.
12562 Frame 1 Frame 2 Frame 3 Frame 4
12564 11111<- 22222 33333<- 44444
12565 11111 22222<- 33333 44444<-
12566 11111<- 22222 33333<- 44444
12567 11111 22222<- 33333 44444<-
12577 @item interleave_bottom, 5
12578 Interleave the lower field from odd frames with the upper field from
12579 even frames, generating a frame with unchanged height at half frame rate.
12584 Frame 1 Frame 2 Frame 3 Frame 4
12586 11111 22222<- 33333 44444<-
12587 11111<- 22222 33333<- 44444
12588 11111 22222<- 33333 44444<-
12589 11111<- 22222 33333<- 44444
12599 @item interlacex2, 6
12600 Double frame rate with unchanged height. Frames are inserted each
12601 containing the second temporal field from the previous input frame and
12602 the first temporal field from the next input frame. This mode relies on
12603 the top_field_first flag. Useful for interlaced video displays with no
12604 field synchronisation.
12609 Frame 1 Frame 2 Frame 3 Frame 4
12611 11111 22222 33333 44444
12612 11111 22222 33333 44444
12613 11111 22222 33333 44444
12614 11111 22222 33333 44444
12617 11111 22222 22222 33333 33333 44444 44444
12618 11111 11111 22222 22222 33333 33333 44444
12619 11111 22222 22222 33333 33333 44444 44444
12620 11111 11111 22222 22222 33333 33333 44444
12624 Move odd frames into the upper field, even into the lower field,
12625 generating a double height frame at same frame rate.
12629 Frame 1 Frame 2 Frame 3 Frame 4
12631 11111 22222 33333 44444
12632 11111 22222 33333 44444
12633 11111 22222 33333 44444
12634 11111 22222 33333 44444
12637 11111 33333 33333 55555
12638 22222 22222 44444 44444
12639 11111 33333 33333 55555
12640 22222 22222 44444 44444
12641 11111 33333 33333 55555
12642 22222 22222 44444 44444
12643 11111 33333 33333 55555
12644 22222 22222 44444 44444
12649 Numeric values are deprecated but are accepted for backward
12650 compatibility reasons.
12652 Default mode is @code{merge}.
12655 Specify flags influencing the filter process.
12657 Available value for @var{flags} is:
12660 @item low_pass_filter, vlfp
12661 Enable vertical low-pass filtering in the filter.
12662 Vertical low-pass filtering is required when creating an interlaced
12663 destination from a progressive source which contains high-frequency
12664 vertical detail. Filtering will reduce interlace 'twitter' and Moire
12667 Vertical low-pass filtering can only be enabled for @option{mode}
12668 @var{interleave_top} and @var{interleave_bottom}.
12675 Transpose rows with columns in the input video and optionally flip it.
12677 It accepts the following parameters:
12682 Specify the transposition direction.
12684 Can assume the following values:
12686 @item 0, 4, cclock_flip
12687 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
12695 Rotate by 90 degrees clockwise, that is:
12703 Rotate by 90 degrees counterclockwise, that is:
12710 @item 3, 7, clock_flip
12711 Rotate by 90 degrees clockwise and vertically flip, that is:
12719 For values between 4-7, the transposition is only done if the input
12720 video geometry is portrait and not landscape. These values are
12721 deprecated, the @code{passthrough} option should be used instead.
12723 Numerical values are deprecated, and should be dropped in favor of
12724 symbolic constants.
12727 Do not apply the transposition if the input geometry matches the one
12728 specified by the specified value. It accepts the following values:
12731 Always apply transposition.
12733 Preserve portrait geometry (when @var{height} >= @var{width}).
12735 Preserve landscape geometry (when @var{width} >= @var{height}).
12738 Default value is @code{none}.
12741 For example to rotate by 90 degrees clockwise and preserve portrait
12744 transpose=dir=1:passthrough=portrait
12747 The command above can also be specified as:
12749 transpose=1:portrait
12753 Trim the input so that the output contains one continuous subpart of the input.
12755 It accepts the following parameters:
12758 Specify the time of the start of the kept section, i.e. the frame with the
12759 timestamp @var{start} will be the first frame in the output.
12762 Specify the time of the first frame that will be dropped, i.e. the frame
12763 immediately preceding the one with the timestamp @var{end} will be the last
12764 frame in the output.
12767 This is the same as @var{start}, except this option sets the start timestamp
12768 in timebase units instead of seconds.
12771 This is the same as @var{end}, except this option sets the end timestamp
12772 in timebase units instead of seconds.
12775 The maximum duration of the output in seconds.
12778 The number of the first frame that should be passed to the output.
12781 The number of the first frame that should be dropped.
12784 @option{start}, @option{end}, and @option{duration} are expressed as time
12785 duration specifications; see
12786 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
12787 for the accepted syntax.
12789 Note that the first two sets of the start/end options and the @option{duration}
12790 option look at the frame timestamp, while the _frame variants simply count the
12791 frames that pass through the filter. Also note that this filter does not modify
12792 the timestamps. If you wish for the output timestamps to start at zero, insert a
12793 setpts filter after the trim filter.
12795 If multiple start or end options are set, this filter tries to be greedy and
12796 keep all the frames that match at least one of the specified constraints. To keep
12797 only the part that matches all the constraints at once, chain multiple trim
12800 The defaults are such that all the input is kept. So it is possible to set e.g.
12801 just the end values to keep everything before the specified time.
12806 Drop everything except the second minute of input:
12808 ffmpeg -i INPUT -vf trim=60:120
12812 Keep only the first second:
12814 ffmpeg -i INPUT -vf trim=duration=1
12823 Sharpen or blur the input video.
12825 It accepts the following parameters:
12828 @item luma_msize_x, lx
12829 Set the luma matrix horizontal size. It must be an odd integer between
12830 3 and 63. The default value is 5.
12832 @item luma_msize_y, ly
12833 Set the luma matrix vertical size. It must be an odd integer between 3
12834 and 63. The default value is 5.
12836 @item luma_amount, la
12837 Set the luma effect strength. It must be a floating point number, reasonable
12838 values lay between -1.5 and 1.5.
12840 Negative values will blur the input video, while positive values will
12841 sharpen it, a value of zero will disable the effect.
12843 Default value is 1.0.
12845 @item chroma_msize_x, cx
12846 Set the chroma matrix horizontal size. It must be an odd integer
12847 between 3 and 63. The default value is 5.
12849 @item chroma_msize_y, cy
12850 Set the chroma matrix vertical size. It must be an odd integer
12851 between 3 and 63. The default value is 5.
12853 @item chroma_amount, ca
12854 Set the chroma effect strength. It must be a floating point number, reasonable
12855 values lay between -1.5 and 1.5.
12857 Negative values will blur the input video, while positive values will
12858 sharpen it, a value of zero will disable the effect.
12860 Default value is 0.0.
12863 If set to 1, specify using OpenCL capabilities, only available if
12864 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
12868 All parameters are optional and default to the equivalent of the
12869 string '5:5:1.0:5:5:0.0'.
12871 @subsection Examples
12875 Apply strong luma sharpen effect:
12877 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
12881 Apply a strong blur of both luma and chroma parameters:
12883 unsharp=7:7:-2:7:7:-2
12889 Apply ultra slow/simple postprocessing filter that compresses and decompresses
12890 the image at several (or - in the case of @option{quality} level @code{8} - all)
12891 shifts and average the results.
12893 The way this differs from the behavior of spp is that uspp actually encodes &
12894 decodes each case with libavcodec Snow, whereas spp uses a simplified intra only 8x8
12895 DCT similar to MJPEG.
12897 The filter accepts the following options:
12901 Set quality. This option defines the number of levels for averaging. It accepts
12902 an integer in the range 0-8. If set to @code{0}, the filter will have no
12903 effect. A value of @code{8} means the higher quality. For each increment of
12904 that value the speed drops by a factor of approximately 2. Default value is
12908 Force a constant quantization parameter. If not set, the filter will use the QP
12909 from the video stream (if available).
12912 @section vectorscope
12914 Display 2 color component values in the two dimensional graph (which is called
12917 This filter accepts the following options:
12921 Set vectorscope mode.
12923 It accepts the following values:
12926 Gray values are displayed on graph, higher brightness means more pixels have
12927 same component color value on location in graph. This is the default mode.
12930 Gray values are displayed on graph. Surrounding pixels values which are not
12931 present in video frame are drawn in gradient of 2 color components which are
12932 set by option @code{x} and @code{y}. The 3rd color component is static.
12935 Actual color components values present in video frame are displayed on graph.
12938 Similar as color2 but higher frequency of same values @code{x} and @code{y}
12939 on graph increases value of another color component, which is luminance by
12940 default values of @code{x} and @code{y}.
12943 Actual colors present in video frame are displayed on graph. If two different
12944 colors map to same position on graph then color with higher value of component
12945 not present in graph is picked.
12948 Gray values are displayed on graph. Similar to @code{color} but with 3rd color
12949 component picked from radial gradient.
12953 Set which color component will be represented on X-axis. Default is @code{1}.
12956 Set which color component will be represented on Y-axis. Default is @code{2}.
12959 Set intensity, used by modes: gray, color, color3 and color5 for increasing brightness
12960 of color component which represents frequency of (X, Y) location in graph.
12965 No envelope, this is default.
12968 Instant envelope, even darkest single pixel will be clearly highlighted.
12971 Hold maximum and minimum values presented in graph over time. This way you
12972 can still spot out of range values without constantly looking at vectorscope.
12975 Peak and instant envelope combined together.
12979 Set what kind of graticule to draw.
12987 Set graticule opacity.
12990 Set graticule flags.
12994 Draw graticule for white point.
12997 Draw graticule for black point.
13000 Draw color points short names.
13004 Set background opacity.
13006 @item lthreshold, l
13007 Set low threshold for color component not represented on X or Y axis.
13008 Values lower than this value will be ignored. Default is 0.
13009 Note this value is multiplied with actual max possible value one pixel component
13010 can have. So for 8-bit input and low threshold value of 0.1 actual threshold
13013 @item hthreshold, h
13014 Set high threshold for color component not represented on X or Y axis.
13015 Values higher than this value will be ignored. Default is 1.
13016 Note this value is multiplied with actual max possible value one pixel component
13017 can have. So for 8-bit input and high threshold value of 0.9 actual threshold
13018 is 0.9 * 255 = 230.
13020 @item colorspace, c
13021 Set what kind of colorspace to use when drawing graticule.
13030 @anchor{vidstabdetect}
13031 @section vidstabdetect
13033 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
13034 @ref{vidstabtransform} for pass 2.
13036 This filter generates a file with relative translation and rotation
13037 transform information about subsequent frames, which is then used by
13038 the @ref{vidstabtransform} filter.
13040 To enable compilation of this filter you need to configure FFmpeg with
13041 @code{--enable-libvidstab}.
13043 This filter accepts the following options:
13047 Set the path to the file used to write the transforms information.
13048 Default value is @file{transforms.trf}.
13051 Set how shaky the video is and how quick the camera is. It accepts an
13052 integer in the range 1-10, a value of 1 means little shakiness, a
13053 value of 10 means strong shakiness. Default value is 5.
13056 Set the accuracy of the detection process. It must be a value in the
13057 range 1-15. A value of 1 means low accuracy, a value of 15 means high
13058 accuracy. Default value is 15.
13061 Set stepsize of the search process. The region around minimum is
13062 scanned with 1 pixel resolution. Default value is 6.
13065 Set minimum contrast. Below this value a local measurement field is
13066 discarded. Must be a floating point value in the range 0-1. Default
13070 Set reference frame number for tripod mode.
13072 If enabled, the motion of the frames is compared to a reference frame
13073 in the filtered stream, identified by the specified number. The idea
13074 is to compensate all movements in a more-or-less static scene and keep
13075 the camera view absolutely still.
13077 If set to 0, it is disabled. The frames are counted starting from 1.
13080 Show fields and transforms in the resulting frames. It accepts an
13081 integer in the range 0-2. Default value is 0, which disables any
13085 @subsection Examples
13089 Use default values:
13095 Analyze strongly shaky movie and put the results in file
13096 @file{mytransforms.trf}:
13098 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
13102 Visualize the result of internal transformations in the resulting
13105 vidstabdetect=show=1
13109 Analyze a video with medium shakiness using @command{ffmpeg}:
13111 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
13115 @anchor{vidstabtransform}
13116 @section vidstabtransform
13118 Video stabilization/deshaking: pass 2 of 2,
13119 see @ref{vidstabdetect} for pass 1.
13121 Read a file with transform information for each frame and
13122 apply/compensate them. Together with the @ref{vidstabdetect}
13123 filter this can be used to deshake videos. See also
13124 @url{http://public.hronopik.de/vid.stab}. It is important to also use
13125 the @ref{unsharp} filter, see below.
13127 To enable compilation of this filter you need to configure FFmpeg with
13128 @code{--enable-libvidstab}.
13130 @subsection Options
13134 Set path to the file used to read the transforms. Default value is
13135 @file{transforms.trf}.
13138 Set the number of frames (value*2 + 1) used for lowpass filtering the
13139 camera movements. Default value is 10.
13141 For example a number of 10 means that 21 frames are used (10 in the
13142 past and 10 in the future) to smoothen the motion in the video. A
13143 larger value leads to a smoother video, but limits the acceleration of
13144 the camera (pan/tilt movements). 0 is a special case where a static
13145 camera is simulated.
13148 Set the camera path optimization algorithm.
13150 Accepted values are:
13153 gaussian kernel low-pass filter on camera motion (default)
13155 averaging on transformations
13159 Set maximal number of pixels to translate frames. Default value is -1,
13163 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
13164 value is -1, meaning no limit.
13167 Specify how to deal with borders that may be visible due to movement
13170 Available values are:
13173 keep image information from previous frame (default)
13175 fill the border black
13179 Invert transforms if set to 1. Default value is 0.
13182 Consider transforms as relative to previous frame if set to 1,
13183 absolute if set to 0. Default value is 0.
13186 Set percentage to zoom. A positive value will result in a zoom-in
13187 effect, a negative value in a zoom-out effect. Default value is 0 (no
13191 Set optimal zooming to avoid borders.
13193 Accepted values are:
13198 optimal static zoom value is determined (only very strong movements
13199 will lead to visible borders) (default)
13201 optimal adaptive zoom value is determined (no borders will be
13202 visible), see @option{zoomspeed}
13205 Note that the value given at zoom is added to the one calculated here.
13208 Set percent to zoom maximally each frame (enabled when
13209 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
13213 Specify type of interpolation.
13215 Available values are:
13220 linear only horizontal
13222 linear in both directions (default)
13224 cubic in both directions (slow)
13228 Enable virtual tripod mode if set to 1, which is equivalent to
13229 @code{relative=0:smoothing=0}. Default value is 0.
13231 Use also @code{tripod} option of @ref{vidstabdetect}.
13234 Increase log verbosity if set to 1. Also the detected global motions
13235 are written to the temporary file @file{global_motions.trf}. Default
13239 @subsection Examples
13243 Use @command{ffmpeg} for a typical stabilization with default values:
13245 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
13248 Note the use of the @ref{unsharp} filter which is always recommended.
13251 Zoom in a bit more and load transform data from a given file:
13253 vidstabtransform=zoom=5:input="mytransforms.trf"
13257 Smoothen the video even more:
13259 vidstabtransform=smoothing=30
13265 Flip the input video vertically.
13267 For example, to vertically flip a video with @command{ffmpeg}:
13269 ffmpeg -i in.avi -vf "vflip" out.avi
13275 Make or reverse a natural vignetting effect.
13277 The filter accepts the following options:
13281 Set lens angle expression as a number of radians.
13283 The value is clipped in the @code{[0,PI/2]} range.
13285 Default value: @code{"PI/5"}
13289 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
13293 Set forward/backward mode.
13295 Available modes are:
13298 The larger the distance from the central point, the darker the image becomes.
13301 The larger the distance from the central point, the brighter the image becomes.
13302 This can be used to reverse a vignette effect, though there is no automatic
13303 detection to extract the lens @option{angle} and other settings (yet). It can
13304 also be used to create a burning effect.
13307 Default value is @samp{forward}.
13310 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
13312 It accepts the following values:
13315 Evaluate expressions only once during the filter initialization.
13318 Evaluate expressions for each incoming frame. This is way slower than the
13319 @samp{init} mode since it requires all the scalers to be re-computed, but it
13320 allows advanced dynamic expressions.
13323 Default value is @samp{init}.
13326 Set dithering to reduce the circular banding effects. Default is @code{1}
13330 Set vignette aspect. This setting allows one to adjust the shape of the vignette.
13331 Setting this value to the SAR of the input will make a rectangular vignetting
13332 following the dimensions of the video.
13334 Default is @code{1/1}.
13337 @subsection Expressions
13339 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
13340 following parameters.
13345 input width and height
13348 the number of input frame, starting from 0
13351 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
13352 @var{TB} units, NAN if undefined
13355 frame rate of the input video, NAN if the input frame rate is unknown
13358 the PTS (Presentation TimeStamp) of the filtered video frame,
13359 expressed in seconds, NAN if undefined
13362 time base of the input video
13366 @subsection Examples
13370 Apply simple strong vignetting effect:
13376 Make a flickering vignetting:
13378 vignette='PI/4+random(1)*PI/50':eval=frame
13384 Stack input videos vertically.
13386 All streams must be of same pixel format and of same width.
13388 Note that this filter is faster than using @ref{overlay} and @ref{pad} filter
13389 to create same output.
13391 The filter accept the following option:
13395 Set number of input streams. Default is 2.
13398 If set to 1, force the output to terminate when the shortest input
13399 terminates. Default value is 0.
13404 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
13405 Deinterlacing Filter").
13407 Based on the process described by Martin Weston for BBC R&D, and
13408 implemented based on the de-interlace algorithm written by Jim
13409 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
13410 uses filter coefficients calculated by BBC R&D.
13412 There are two sets of filter coefficients, so called "simple":
13413 and "complex". Which set of filter coefficients is used can
13414 be set by passing an optional parameter:
13418 Set the interlacing filter coefficients. Accepts one of the following values:
13422 Simple filter coefficient set.
13424 More-complex filter coefficient set.
13426 Default value is @samp{complex}.
13429 Specify which frames to deinterlace. Accept one of the following values:
13433 Deinterlace all frames,
13435 Only deinterlace frames marked as interlaced.
13438 Default value is @samp{all}.
13442 Video waveform monitor.
13444 The waveform monitor plots color component intensity. By default luminance
13445 only. Each column of the waveform corresponds to a column of pixels in the
13448 It accepts the following options:
13452 Can be either @code{row}, or @code{column}. Default is @code{column}.
13453 In row mode, the graph on the left side represents color component value 0 and
13454 the right side represents value = 255. In column mode, the top side represents
13455 color component value = 0 and bottom side represents value = 255.
13458 Set intensity. Smaller values are useful to find out how many values of the same
13459 luminance are distributed across input rows/columns.
13460 Default value is @code{0.04}. Allowed range is [0, 1].
13463 Set mirroring mode. @code{0} means unmirrored, @code{1} means mirrored.
13464 In mirrored mode, higher values will be represented on the left
13465 side for @code{row} mode and at the top for @code{column} mode. Default is
13466 @code{1} (mirrored).
13470 It accepts the following values:
13473 Presents information identical to that in the @code{parade}, except
13474 that the graphs representing color components are superimposed directly
13477 This display mode makes it easier to spot relative differences or similarities
13478 in overlapping areas of the color components that are supposed to be identical,
13479 such as neutral whites, grays, or blacks.
13482 Display separate graph for the color components side by side in
13483 @code{row} mode or one below the other in @code{column} mode.
13486 Display separate graph for the color components side by side in
13487 @code{column} mode or one below the other in @code{row} mode.
13489 Using this display mode makes it easy to spot color casts in the highlights
13490 and shadows of an image, by comparing the contours of the top and the bottom
13491 graphs of each waveform. Since whites, grays, and blacks are characterized
13492 by exactly equal amounts of red, green, and blue, neutral areas of the picture
13493 should display three waveforms of roughly equal width/height. If not, the
13494 correction is easy to perform by making level adjustments the three waveforms.
13496 Default is @code{stack}.
13498 @item components, c
13499 Set which color components to display. Default is 1, which means only luminance
13500 or red color component if input is in RGB colorspace. If is set for example to
13501 7 it will display all 3 (if) available color components.
13506 No envelope, this is default.
13509 Instant envelope, minimum and maximum values presented in graph will be easily
13510 visible even with small @code{step} value.
13513 Hold minimum and maximum values presented in graph across time. This way you
13514 can still spot out of range values without constantly looking at waveforms.
13517 Peak and instant envelope combined together.
13523 No filtering, this is default.
13526 Luma and chroma combined together.
13529 Similar as above, but shows difference between blue and red chroma.
13532 Displays only chroma.
13535 Displays actual color value on waveform.
13538 Similar as above, but with luma showing frequency of chroma values.
13542 Set which graticule to display.
13546 Do not display graticule.
13549 Display green graticule showing legal broadcast ranges.
13553 Set graticule opacity.
13556 Set graticule flags.
13560 Draw numbers above lines. By default enabled.
13563 Draw dots instead of lines.
13567 Set scale used for displaying graticule.
13574 Default is digital.
13578 Apply the xBR high-quality magnification filter which is designed for pixel
13579 art. It follows a set of edge-detection rules, see
13580 @url{http://www.libretro.com/forums/viewtopic.php?f=6&t=134}.
13582 It accepts the following option:
13586 Set the scaling dimension: @code{2} for @code{2xBR}, @code{3} for
13587 @code{3xBR} and @code{4} for @code{4xBR}.
13588 Default is @code{3}.
13594 Deinterlace the input video ("yadif" means "yet another deinterlacing
13597 It accepts the following parameters:
13603 The interlacing mode to adopt. It accepts one of the following values:
13606 @item 0, send_frame
13607 Output one frame for each frame.
13608 @item 1, send_field
13609 Output one frame for each field.
13610 @item 2, send_frame_nospatial
13611 Like @code{send_frame}, but it skips the spatial interlacing check.
13612 @item 3, send_field_nospatial
13613 Like @code{send_field}, but it skips the spatial interlacing check.
13616 The default value is @code{send_frame}.
13619 The picture field parity assumed for the input interlaced video. It accepts one
13620 of the following values:
13624 Assume the top field is first.
13626 Assume the bottom field is first.
13628 Enable automatic detection of field parity.
13631 The default value is @code{auto}.
13632 If the interlacing is unknown or the decoder does not export this information,
13633 top field first will be assumed.
13636 Specify which frames to deinterlace. Accept one of the following
13641 Deinterlace all frames.
13642 @item 1, interlaced
13643 Only deinterlace frames marked as interlaced.
13646 The default value is @code{all}.
13651 Apply Zoom & Pan effect.
13653 This filter accepts the following options:
13657 Set the zoom expression. Default is 1.
13661 Set the x and y expression. Default is 0.
13664 Set the duration expression in number of frames.
13665 This sets for how many number of frames effect will last for
13666 single input image.
13669 Set the output image size, default is 'hd720'.
13672 Set the output frame rate, default is '25'.
13675 Each expression can contain the following constants:
13694 Output frame count.
13698 Last calculated 'x' and 'y' position from 'x' and 'y' expression
13699 for current input frame.
13703 'x' and 'y' of last output frame of previous input frame or 0 when there was
13704 not yet such frame (first input frame).
13707 Last calculated zoom from 'z' expression for current input frame.
13710 Last calculated zoom of last output frame of previous input frame.
13713 Number of output frames for current input frame. Calculated from 'd' expression
13714 for each input frame.
13717 number of output frames created for previous input frame
13720 Rational number: input width / input height
13723 sample aspect ratio
13726 display aspect ratio
13730 @subsection Examples
13734 Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
13736 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
13740 Zoom-in up to 1.5 and pan always at center of picture:
13742 zoompan=z='min(zoom+0.0015,1.5)':d=700:x='iw/2-(iw/zoom/2)':y='ih/2-(ih/zoom/2)'
13747 Scale (resize) the input video, using the z.lib library:
13748 https://github.com/sekrit-twc/zimg.
13750 The zscale filter forces the output display aspect ratio to be the same
13751 as the input, by changing the output sample aspect ratio.
13753 If the input image format is different from the format requested by
13754 the next filter, the zscale filter will convert the input to the
13757 @subsection Options
13758 The filter accepts the following options.
13763 Set the output video dimension expression. Default value is the input
13766 If the @var{width} or @var{w} is 0, the input width is used for the output.
13767 If the @var{height} or @var{h} is 0, the input height is used for the output.
13769 If one of the values is -1, the zscale filter will use a value that
13770 maintains the aspect ratio of the input image, calculated from the
13771 other specified dimension. If both of them are -1, the input size is
13774 If one of the values is -n with n > 1, the zscale filter will also use a value
13775 that maintains the aspect ratio of the input image, calculated from the other
13776 specified dimension. After that it will, however, make sure that the calculated
13777 dimension is divisible by n and adjust the value if necessary.
13779 See below for the list of accepted constants for use in the dimension
13783 Set the video size. For the syntax of this option, check the
13784 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13787 Set the dither type.
13789 Possible values are:
13794 @item error_diffusion
13800 Set the resize filter type.
13802 Possible values are:
13812 Default is bilinear.
13815 Set the color range.
13817 Possible values are:
13824 Default is same as input.
13827 Set the color primaries.
13829 Possible values are:
13839 Default is same as input.
13842 Set the transfer characteristics.
13844 Possible values are:
13855 Default is same as input.
13858 Set the colorspace matrix.
13860 Possible value are:
13871 Default is same as input.
13874 Set the input color range.
13876 Possible values are:
13883 Default is same as input.
13885 @item primariesin, pin
13886 Set the input color primaries.
13888 Possible values are:
13898 Default is same as input.
13900 @item transferin, tin
13901 Set the input transfer characteristics.
13903 Possible values are:
13914 Default is same as input.
13916 @item matrixin, min
13917 Set the input colorspace matrix.
13919 Possible value are:
13931 The values of the @option{w} and @option{h} options are expressions
13932 containing the following constants:
13937 The input width and height
13941 These are the same as @var{in_w} and @var{in_h}.
13945 The output (scaled) width and height
13949 These are the same as @var{out_w} and @var{out_h}
13952 The same as @var{iw} / @var{ih}
13955 input sample aspect ratio
13958 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
13962 horizontal and vertical input chroma subsample values. For example for the
13963 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
13967 horizontal and vertical output chroma subsample values. For example for the
13968 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
13974 @c man end VIDEO FILTERS
13976 @chapter Video Sources
13977 @c man begin VIDEO SOURCES
13979 Below is a description of the currently available video sources.
13983 Buffer video frames, and make them available to the filter chain.
13985 This source is mainly intended for a programmatic use, in particular
13986 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
13988 It accepts the following parameters:
13993 Specify the size (width and height) of the buffered video frames. For the
13994 syntax of this option, check the
13995 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
13998 The input video width.
14001 The input video height.
14004 A string representing the pixel format of the buffered video frames.
14005 It may be a number corresponding to a pixel format, or a pixel format
14009 Specify the timebase assumed by the timestamps of the buffered frames.
14012 Specify the frame rate expected for the video stream.
14014 @item pixel_aspect, sar
14015 The sample (pixel) aspect ratio of the input video.
14018 Specify the optional parameters to be used for the scale filter which
14019 is automatically inserted when an input change is detected in the
14020 input size or format.
14022 @item hw_frames_ctx
14023 When using a hardware pixel format, this should be a reference to an
14024 AVHWFramesContext describing input frames.
14029 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
14032 will instruct the source to accept video frames with size 320x240 and
14033 with format "yuv410p", assuming 1/24 as the timestamps timebase and
14034 square pixels (1:1 sample aspect ratio).
14035 Since the pixel format with name "yuv410p" corresponds to the number 6
14036 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
14037 this example corresponds to:
14039 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
14042 Alternatively, the options can be specified as a flat string, but this
14043 syntax is deprecated:
14045 @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}]
14049 Create a pattern generated by an elementary cellular automaton.
14051 The initial state of the cellular automaton can be defined through the
14052 @option{filename}, and @option{pattern} options. If such options are
14053 not specified an initial state is created randomly.
14055 At each new frame a new row in the video is filled with the result of
14056 the cellular automaton next generation. The behavior when the whole
14057 frame is filled is defined by the @option{scroll} option.
14059 This source accepts the following options:
14063 Read the initial cellular automaton state, i.e. the starting row, from
14064 the specified file.
14065 In the file, each non-whitespace character is considered an alive
14066 cell, a newline will terminate the row, and further characters in the
14067 file will be ignored.
14070 Read the initial cellular automaton state, i.e. the starting row, from
14071 the specified string.
14073 Each non-whitespace character in the string is considered an alive
14074 cell, a newline will terminate the row, and further characters in the
14075 string will be ignored.
14078 Set the video rate, that is the number of frames generated per second.
14081 @item random_fill_ratio, ratio
14082 Set the random fill ratio for the initial cellular automaton row. It
14083 is a floating point number value ranging from 0 to 1, defaults to
14086 This option is ignored when a file or a pattern is specified.
14088 @item random_seed, seed
14089 Set the seed for filling randomly the initial row, must be an integer
14090 included between 0 and UINT32_MAX. If not specified, or if explicitly
14091 set to -1, the filter will try to use a good random seed on a best
14095 Set the cellular automaton rule, it is a number ranging from 0 to 255.
14096 Default value is 110.
14099 Set the size of the output video. For the syntax of this option, check the
14100 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14102 If @option{filename} or @option{pattern} is specified, the size is set
14103 by default to the width of the specified initial state row, and the
14104 height is set to @var{width} * PHI.
14106 If @option{size} is set, it must contain the width of the specified
14107 pattern string, and the specified pattern will be centered in the
14110 If a filename or a pattern string is not specified, the size value
14111 defaults to "320x518" (used for a randomly generated initial state).
14114 If set to 1, scroll the output upward when all the rows in the output
14115 have been already filled. If set to 0, the new generated row will be
14116 written over the top row just after the bottom row is filled.
14119 @item start_full, full
14120 If set to 1, completely fill the output with generated rows before
14121 outputting the first frame.
14122 This is the default behavior, for disabling set the value to 0.
14125 If set to 1, stitch the left and right row edges together.
14126 This is the default behavior, for disabling set the value to 0.
14129 @subsection Examples
14133 Read the initial state from @file{pattern}, and specify an output of
14136 cellauto=f=pattern:s=200x400
14140 Generate a random initial row with a width of 200 cells, with a fill
14143 cellauto=ratio=2/3:s=200x200
14147 Create a pattern generated by rule 18 starting by a single alive cell
14148 centered on an initial row with width 100:
14150 cellauto=p=@@:s=100x400:full=0:rule=18
14154 Specify a more elaborated initial pattern:
14156 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
14161 @anchor{coreimagesrc}
14162 @section coreimagesrc
14163 Video source generated on GPU using Apple's CoreImage API on OSX.
14165 This video source is a specialized version of the @ref{coreimage} video filter.
14166 Use a core image generator at the beginning of the applied filterchain to
14167 generate the content.
14169 The coreimagesrc video source accepts the following options:
14171 @item list_generators
14172 List all available generators along with all their respective options as well as
14173 possible minimum and maximum values along with the default values.
14175 list_generators=true
14179 Specify the size of the sourced video. For the syntax of this option, check the
14180 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14181 The default value is @code{320x240}.
14184 Specify the frame rate of the sourced video, as the number of frames
14185 generated per second. It has to be a string in the format
14186 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
14187 number or a valid video frame rate abbreviation. The default value is
14191 Set the sample aspect ratio of the sourced video.
14194 Set the duration of the sourced video. See
14195 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
14196 for the accepted syntax.
14198 If not specified, or the expressed duration is negative, the video is
14199 supposed to be generated forever.
14202 Additionally, all options of the @ref{coreimage} video filter are accepted.
14203 A complete filterchain can be used for further processing of the
14204 generated input without CPU-HOST transfer. See @ref{coreimage} documentation
14205 and examples for details.
14207 @subsection Examples
14212 Use CIQRCodeGenerator to create a QR code for the FFmpeg homepage,
14213 given as complete and escaped command-line for Apple's standard bash shell:
14215 ffmpeg -f lavfi -i coreimagesrc=s=100x100:filter=CIQRCodeGenerator@@inputMessage=https\\\\\://FFmpeg.org/@@inputCorrectionLevel=H -frames:v 1 QRCode.png
14217 This example is equivalent to the QRCode example of @ref{coreimage} without the
14218 need for a nullsrc video source.
14222 @section mandelbrot
14224 Generate a Mandelbrot set fractal, and progressively zoom towards the
14225 point specified with @var{start_x} and @var{start_y}.
14227 This source accepts the following options:
14232 Set the terminal pts value. Default value is 400.
14235 Set the terminal scale value.
14236 Must be a floating point value. Default value is 0.3.
14239 Set the inner coloring mode, that is the algorithm used to draw the
14240 Mandelbrot fractal internal region.
14242 It shall assume one of the following values:
14247 Show time until convergence.
14249 Set color based on point closest to the origin of the iterations.
14254 Default value is @var{mincol}.
14257 Set the bailout value. Default value is 10.0.
14260 Set the maximum of iterations performed by the rendering
14261 algorithm. Default value is 7189.
14264 Set outer coloring mode.
14265 It shall assume one of following values:
14267 @item iteration_count
14268 Set iteration cound mode.
14269 @item normalized_iteration_count
14270 set normalized iteration count mode.
14272 Default value is @var{normalized_iteration_count}.
14275 Set frame rate, expressed as number of frames per second. Default
14279 Set frame size. For the syntax of this option, check the "Video
14280 size" section in the ffmpeg-utils manual. Default value is "640x480".
14283 Set the initial scale value. Default value is 3.0.
14286 Set the initial x position. Must be a floating point value between
14287 -100 and 100. Default value is -0.743643887037158704752191506114774.
14290 Set the initial y position. Must be a floating point value between
14291 -100 and 100. Default value is -0.131825904205311970493132056385139.
14296 Generate various test patterns, as generated by the MPlayer test filter.
14298 The size of the generated video is fixed, and is 256x256.
14299 This source is useful in particular for testing encoding features.
14301 This source accepts the following options:
14306 Specify the frame rate of the sourced video, as the number of frames
14307 generated per second. It has to be a string in the format
14308 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
14309 number or a valid video frame rate abbreviation. The default value is
14313 Set the duration of the sourced video. See
14314 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
14315 for the accepted syntax.
14317 If not specified, or the expressed duration is negative, the video is
14318 supposed to be generated forever.
14322 Set the number or the name of the test to perform. Supported tests are:
14338 Default value is "all", which will cycle through the list of all tests.
14343 mptestsrc=t=dc_luma
14346 will generate a "dc_luma" test pattern.
14348 @section frei0r_src
14350 Provide a frei0r source.
14352 To enable compilation of this filter you need to install the frei0r
14353 header and configure FFmpeg with @code{--enable-frei0r}.
14355 This source accepts the following parameters:
14360 The size of the video to generate. For the syntax of this option, check the
14361 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14364 The framerate of the generated video. It may be a string of the form
14365 @var{num}/@var{den} or a frame rate abbreviation.
14368 The name to the frei0r source to load. For more information regarding frei0r and
14369 how to set the parameters, read the @ref{frei0r} section in the video filters
14372 @item filter_params
14373 A '|'-separated list of parameters to pass to the frei0r source.
14377 For example, to generate a frei0r partik0l source with size 200x200
14378 and frame rate 10 which is overlaid on the overlay filter main input:
14380 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
14385 Generate a life pattern.
14387 This source is based on a generalization of John Conway's life game.
14389 The sourced input represents a life grid, each pixel represents a cell
14390 which can be in one of two possible states, alive or dead. Every cell
14391 interacts with its eight neighbours, which are the cells that are
14392 horizontally, vertically, or diagonally adjacent.
14394 At each interaction the grid evolves according to the adopted rule,
14395 which specifies the number of neighbor alive cells which will make a
14396 cell stay alive or born. The @option{rule} option allows one to specify
14399 This source accepts the following options:
14403 Set the file from which to read the initial grid state. In the file,
14404 each non-whitespace character is considered an alive cell, and newline
14405 is used to delimit the end of each row.
14407 If this option is not specified, the initial grid is generated
14411 Set the video rate, that is the number of frames generated per second.
14414 @item random_fill_ratio, ratio
14415 Set the random fill ratio for the initial random grid. It is a
14416 floating point number value ranging from 0 to 1, defaults to 1/PHI.
14417 It is ignored when a file is specified.
14419 @item random_seed, seed
14420 Set the seed for filling the initial random grid, must be an integer
14421 included between 0 and UINT32_MAX. If not specified, or if explicitly
14422 set to -1, the filter will try to use a good random seed on a best
14428 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
14429 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
14430 @var{NS} specifies the number of alive neighbor cells which make a
14431 live cell stay alive, and @var{NB} the number of alive neighbor cells
14432 which make a dead cell to become alive (i.e. to "born").
14433 "s" and "b" can be used in place of "S" and "B", respectively.
14435 Alternatively a rule can be specified by an 18-bits integer. The 9
14436 high order bits are used to encode the next cell state if it is alive
14437 for each number of neighbor alive cells, the low order bits specify
14438 the rule for "borning" new cells. Higher order bits encode for an
14439 higher number of neighbor cells.
14440 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
14441 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
14443 Default value is "S23/B3", which is the original Conway's game of life
14444 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
14445 cells, and will born a new cell if there are three alive cells around
14449 Set the size of the output video. For the syntax of this option, check the
14450 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14452 If @option{filename} is specified, the size is set by default to the
14453 same size of the input file. If @option{size} is set, it must contain
14454 the size specified in the input file, and the initial grid defined in
14455 that file is centered in the larger resulting area.
14457 If a filename is not specified, the size value defaults to "320x240"
14458 (used for a randomly generated initial grid).
14461 If set to 1, stitch the left and right grid edges together, and the
14462 top and bottom edges also. Defaults to 1.
14465 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
14466 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
14467 value from 0 to 255.
14470 Set the color of living (or new born) cells.
14473 Set the color of dead cells. If @option{mold} is set, this is the first color
14474 used to represent a dead cell.
14477 Set mold color, for definitely dead and moldy cells.
14479 For the syntax of these 3 color options, check the "Color" section in the
14480 ffmpeg-utils manual.
14483 @subsection Examples
14487 Read a grid from @file{pattern}, and center it on a grid of size
14490 life=f=pattern:s=300x300
14494 Generate a random grid of size 200x200, with a fill ratio of 2/3:
14496 life=ratio=2/3:s=200x200
14500 Specify a custom rule for evolving a randomly generated grid:
14506 Full example with slow death effect (mold) using @command{ffplay}:
14508 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
14515 @anchor{haldclutsrc}
14517 @anchor{rgbtestsrc}
14519 @anchor{smptehdbars}
14522 @section allrgb, allyuv, color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc, testsrc2
14524 The @code{allrgb} source returns frames of size 4096x4096 of all rgb colors.
14526 The @code{allyuv} source returns frames of size 4096x4096 of all yuv colors.
14528 The @code{color} source provides an uniformly colored input.
14530 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
14531 @ref{haldclut} filter.
14533 The @code{nullsrc} source returns unprocessed video frames. It is
14534 mainly useful to be employed in analysis / debugging tools, or as the
14535 source for filters which ignore the input data.
14537 The @code{rgbtestsrc} source generates an RGB test pattern useful for
14538 detecting RGB vs BGR issues. You should see a red, green and blue
14539 stripe from top to bottom.
14541 The @code{smptebars} source generates a color bars pattern, based on
14542 the SMPTE Engineering Guideline EG 1-1990.
14544 The @code{smptehdbars} source generates a color bars pattern, based on
14545 the SMPTE RP 219-2002.
14547 The @code{testsrc} source generates a test video pattern, showing a
14548 color pattern, a scrolling gradient and a timestamp. This is mainly
14549 intended for testing purposes.
14551 The @code{testsrc2} source is similar to testsrc, but supports more
14552 pixel formats instead of just @code{rgb24}. This allows using it as an
14553 input for other tests without requiring a format conversion.
14555 The sources accept the following parameters:
14560 Specify the color of the source, only available in the @code{color}
14561 source. For the syntax of this option, check the "Color" section in the
14562 ffmpeg-utils manual.
14565 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
14566 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
14567 pixels to be used as identity matrix for 3D lookup tables. Each component is
14568 coded on a @code{1/(N*N)} scale.
14571 Specify the size of the sourced video. For the syntax of this option, check the
14572 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14573 The default value is @code{320x240}.
14575 This option is not available with the @code{haldclutsrc} filter.
14578 Specify the frame rate of the sourced video, as the number of frames
14579 generated per second. It has to be a string in the format
14580 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
14581 number or a valid video frame rate abbreviation. The default value is
14585 Set the sample aspect ratio of the sourced video.
14588 Set the duration of the sourced video. See
14589 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
14590 for the accepted syntax.
14592 If not specified, or the expressed duration is negative, the video is
14593 supposed to be generated forever.
14596 Set the number of decimals to show in the timestamp, only available in the
14597 @code{testsrc} source.
14599 The displayed timestamp value will correspond to the original
14600 timestamp value multiplied by the power of 10 of the specified
14601 value. Default value is 0.
14604 For example the following:
14606 testsrc=duration=5.3:size=qcif:rate=10
14609 will generate a video with a duration of 5.3 seconds, with size
14610 176x144 and a frame rate of 10 frames per second.
14612 The following graph description will generate a red source
14613 with an opacity of 0.2, with size "qcif" and a frame rate of 10
14616 color=c=red@@0.2:s=qcif:r=10
14619 If the input content is to be ignored, @code{nullsrc} can be used. The
14620 following command generates noise in the luminance plane by employing
14621 the @code{geq} filter:
14623 nullsrc=s=256x256, geq=random(1)*255:128:128
14626 @subsection Commands
14628 The @code{color} source supports the following commands:
14632 Set the color of the created image. Accepts the same syntax of the
14633 corresponding @option{color} option.
14636 @c man end VIDEO SOURCES
14638 @chapter Video Sinks
14639 @c man begin VIDEO SINKS
14641 Below is a description of the currently available video sinks.
14643 @section buffersink
14645 Buffer video frames, and make them available to the end of the filter
14648 This sink is mainly intended for programmatic use, in particular
14649 through the interface defined in @file{libavfilter/buffersink.h}
14650 or the options system.
14652 It accepts a pointer to an AVBufferSinkContext structure, which
14653 defines the incoming buffers' formats, to be passed as the opaque
14654 parameter to @code{avfilter_init_filter} for initialization.
14658 Null video sink: do absolutely nothing with the input video. It is
14659 mainly useful as a template and for use in analysis / debugging
14662 @c man end VIDEO SINKS
14664 @chapter Multimedia Filters
14665 @c man begin MULTIMEDIA FILTERS
14667 Below is a description of the currently available multimedia filters.
14669 @section ahistogram
14671 Convert input audio to a video output, displaying the volume histogram.
14673 The filter accepts the following options:
14677 Specify how histogram is calculated.
14679 It accepts the following values:
14682 Use single histogram for all channels.
14684 Use separate histogram for each channel.
14686 Default is @code{single}.
14689 Set frame rate, expressed as number of frames per second. Default
14693 Specify the video size for the output. For the syntax of this option, check the
14694 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14695 Default value is @code{hd720}.
14700 It accepts the following values:
14711 reverse logarithmic
14713 Default is @code{log}.
14716 Set amplitude scale.
14718 It accepts the following values:
14725 Default is @code{log}.
14728 Set how much frames to accumulate in histogram.
14729 Defauls is 1. Setting this to -1 accumulates all frames.
14732 Set histogram ratio of window height.
14735 Set sonogram sliding.
14737 It accepts the following values:
14740 replace old rows with new ones.
14742 scroll from top to bottom.
14744 Default is @code{replace}.
14747 @section aphasemeter
14749 Convert input audio to a video output, displaying the audio phase.
14751 The filter accepts the following options:
14755 Set the output frame rate. Default value is @code{25}.
14758 Set the video size for the output. For the syntax of this option, check the
14759 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14760 Default value is @code{800x400}.
14765 Specify the red, green, blue contrast. Default values are @code{2},
14766 @code{7} and @code{1}.
14767 Allowed range is @code{[0, 255]}.
14770 Set color which will be used for drawing median phase. If color is
14771 @code{none} which is default, no median phase value will be drawn.
14774 The filter also exports the frame metadata @code{lavfi.aphasemeter.phase} which
14775 represents mean phase of current audio frame. Value is in range @code{[-1, 1]}.
14776 The @code{-1} means left and right channels are completely out of phase and
14777 @code{1} means channels are in phase.
14779 @section avectorscope
14781 Convert input audio to a video output, representing the audio vector
14784 The filter is used to measure the difference between channels of stereo
14785 audio stream. A monoaural signal, consisting of identical left and right
14786 signal, results in straight vertical line. Any stereo separation is visible
14787 as a deviation from this line, creating a Lissajous figure.
14788 If the straight (or deviation from it) but horizontal line appears this
14789 indicates that the left and right channels are out of phase.
14791 The filter accepts the following options:
14795 Set the vectorscope mode.
14797 Available values are:
14800 Lissajous rotated by 45 degrees.
14803 Same as above but not rotated.
14806 Shape resembling half of circle.
14809 Default value is @samp{lissajous}.
14812 Set the video size for the output. For the syntax of this option, check the
14813 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
14814 Default value is @code{400x400}.
14817 Set the output frame rate. Default value is @code{25}.
14823 Specify the red, green, blue and alpha contrast. Default values are @code{40},
14824 @code{160}, @code{80} and @code{255}.
14825 Allowed range is @code{[0, 255]}.
14831 Specify the red, green, blue and alpha fade. Default values are @code{15},
14832 @code{10}, @code{5} and @code{5}.
14833 Allowed range is @code{[0, 255]}.
14836 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
14839 Set the vectorscope drawing mode.
14841 Available values are:
14844 Draw dot for each sample.
14847 Draw line between previous and current sample.
14850 Default value is @samp{dot}.
14853 @subsection Examples
14857 Complete example using @command{ffplay}:
14859 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
14860 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
14864 @section bench, abench
14866 Benchmark part of a filtergraph.
14868 The filter accepts the following options:
14872 Start or stop a timer.
14874 Available values are:
14877 Get the current time, set it as frame metadata (using the key
14878 @code{lavfi.bench.start_time}), and forward the frame to the next filter.
14881 Get the current time and fetch the @code{lavfi.bench.start_time} metadata from
14882 the input frame metadata to get the time difference. Time difference, average,
14883 maximum and minimum time (respectively @code{t}, @code{avg}, @code{max} and
14884 @code{min}) are then printed. The timestamps are expressed in seconds.
14888 @subsection Examples
14892 Benchmark @ref{selectivecolor} filter:
14894 bench=start,selectivecolor=reds=-.2 .12 -.49,bench=stop
14900 Concatenate audio and video streams, joining them together one after the
14903 The filter works on segments of synchronized video and audio streams. All
14904 segments must have the same number of streams of each type, and that will
14905 also be the number of streams at output.
14907 The filter accepts the following options:
14912 Set the number of segments. Default is 2.
14915 Set the number of output video streams, that is also the number of video
14916 streams in each segment. Default is 1.
14919 Set the number of output audio streams, that is also the number of audio
14920 streams in each segment. Default is 0.
14923 Activate unsafe mode: do not fail if segments have a different format.
14927 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
14928 @var{a} audio outputs.
14930 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
14931 segment, in the same order as the outputs, then the inputs for the second
14934 Related streams do not always have exactly the same duration, for various
14935 reasons including codec frame size or sloppy authoring. For that reason,
14936 related synchronized streams (e.g. a video and its audio track) should be
14937 concatenated at once. The concat filter will use the duration of the longest
14938 stream in each segment (except the last one), and if necessary pad shorter
14939 audio streams with silence.
14941 For this filter to work correctly, all segments must start at timestamp 0.
14943 All corresponding streams must have the same parameters in all segments; the
14944 filtering system will automatically select a common pixel format for video
14945 streams, and a common sample format, sample rate and channel layout for
14946 audio streams, but other settings, such as resolution, must be converted
14947 explicitly by the user.
14949 Different frame rates are acceptable but will result in variable frame rate
14950 at output; be sure to configure the output file to handle it.
14952 @subsection Examples
14956 Concatenate an opening, an episode and an ending, all in bilingual version
14957 (video in stream 0, audio in streams 1 and 2):
14959 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
14960 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
14961 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
14962 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
14966 Concatenate two parts, handling audio and video separately, using the
14967 (a)movie sources, and adjusting the resolution:
14969 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
14970 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
14971 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
14973 Note that a desync will happen at the stitch if the audio and video streams
14974 do not have exactly the same duration in the first file.
14981 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
14982 it unchanged. By default, it logs a message at a frequency of 10Hz with the
14983 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
14984 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
14986 The filter also has a video output (see the @var{video} option) with a real
14987 time graph to observe the loudness evolution. The graphic contains the logged
14988 message mentioned above, so it is not printed anymore when this option is set,
14989 unless the verbose logging is set. The main graphing area contains the
14990 short-term loudness (3 seconds of analysis), and the gauge on the right is for
14991 the momentary loudness (400 milliseconds).
14993 More information about the Loudness Recommendation EBU R128 on
14994 @url{http://tech.ebu.ch/loudness}.
14996 The filter accepts the following options:
15001 Activate the video output. The audio stream is passed unchanged whether this
15002 option is set or no. The video stream will be the first output stream if
15003 activated. Default is @code{0}.
15006 Set the video size. This option is for video only. For the syntax of this
15008 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15009 Default and minimum resolution is @code{640x480}.
15012 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
15013 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
15014 other integer value between this range is allowed.
15017 Set metadata injection. If set to @code{1}, the audio input will be segmented
15018 into 100ms output frames, each of them containing various loudness information
15019 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
15021 Default is @code{0}.
15024 Force the frame logging level.
15026 Available values are:
15029 information logging level
15031 verbose logging level
15034 By default, the logging level is set to @var{info}. If the @option{video} or
15035 the @option{metadata} options are set, it switches to @var{verbose}.
15040 Available modes can be cumulated (the option is a @code{flag} type). Possible
15044 Disable any peak mode (default).
15046 Enable sample-peak mode.
15048 Simple peak mode looking for the higher sample value. It logs a message
15049 for sample-peak (identified by @code{SPK}).
15051 Enable true-peak mode.
15053 If enabled, the peak lookup is done on an over-sampled version of the input
15054 stream for better peak accuracy. It logs a message for true-peak.
15055 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
15056 This mode requires a build with @code{libswresample}.
15060 Treat mono input files as "dual mono". If a mono file is intended for playback
15061 on a stereo system, its EBU R128 measurement will be perceptually incorrect.
15062 If set to @code{true}, this option will compensate for this effect.
15063 Multi-channel input files are not affected by this option.
15066 Set a specific pan law to be used for the measurement of dual mono files.
15067 This parameter is optional, and has a default value of -3.01dB.
15070 @subsection Examples
15074 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
15076 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
15080 Run an analysis with @command{ffmpeg}:
15082 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
15086 @section interleave, ainterleave
15088 Temporally interleave frames from several inputs.
15090 @code{interleave} works with video inputs, @code{ainterleave} with audio.
15092 These filters read frames from several inputs and send the oldest
15093 queued frame to the output.
15095 Input streams must have a well defined, monotonically increasing frame
15098 In order to submit one frame to output, these filters need to enqueue
15099 at least one frame for each input, so they cannot work in case one
15100 input is not yet terminated and will not receive incoming frames.
15102 For example consider the case when one input is a @code{select} filter
15103 which always drop input frames. The @code{interleave} filter will keep
15104 reading from that input, but it will never be able to send new frames
15105 to output until the input will send an end-of-stream signal.
15107 Also, depending on inputs synchronization, the filters will drop
15108 frames in case one input receives more frames than the other ones, and
15109 the queue is already filled.
15111 These filters accept the following options:
15115 Set the number of different inputs, it is 2 by default.
15118 @subsection Examples
15122 Interleave frames belonging to different streams using @command{ffmpeg}:
15124 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
15128 Add flickering blur effect:
15130 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
15134 @section perms, aperms
15136 Set read/write permissions for the output frames.
15138 These filters are mainly aimed at developers to test direct path in the
15139 following filter in the filtergraph.
15141 The filters accept the following options:
15145 Select the permissions mode.
15147 It accepts the following values:
15150 Do nothing. This is the default.
15152 Set all the output frames read-only.
15154 Set all the output frames directly writable.
15156 Make the frame read-only if writable, and writable if read-only.
15158 Set each output frame read-only or writable randomly.
15162 Set the seed for the @var{random} mode, must be an integer included between
15163 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
15164 @code{-1}, the filter will try to use a good random seed on a best effort
15168 Note: in case of auto-inserted filter between the permission filter and the
15169 following one, the permission might not be received as expected in that
15170 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
15171 perms/aperms filter can avoid this problem.
15173 @section realtime, arealtime
15175 Slow down filtering to match real time approximatively.
15177 These filters will pause the filtering for a variable amount of time to
15178 match the output rate with the input timestamps.
15179 They are similar to the @option{re} option to @code{ffmpeg}.
15181 They accept the following options:
15185 Time limit for the pauses. Any pause longer than that will be considered
15186 a timestamp discontinuity and reset the timer. Default is 2 seconds.
15189 @section select, aselect
15191 Select frames to pass in output.
15193 This filter accepts the following options:
15198 Set expression, which is evaluated for each input frame.
15200 If the expression is evaluated to zero, the frame is discarded.
15202 If the evaluation result is negative or NaN, the frame is sent to the
15203 first output; otherwise it is sent to the output with index
15204 @code{ceil(val)-1}, assuming that the input index starts from 0.
15206 For example a value of @code{1.2} corresponds to the output with index
15207 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
15210 Set the number of outputs. The output to which to send the selected
15211 frame is based on the result of the evaluation. Default value is 1.
15214 The expression can contain the following constants:
15218 The (sequential) number of the filtered frame, starting from 0.
15221 The (sequential) number of the selected frame, starting from 0.
15223 @item prev_selected_n
15224 The sequential number of the last selected frame. It's NAN if undefined.
15227 The timebase of the input timestamps.
15230 The PTS (Presentation TimeStamp) of the filtered video frame,
15231 expressed in @var{TB} units. It's NAN if undefined.
15234 The PTS of the filtered video frame,
15235 expressed in seconds. It's NAN if undefined.
15238 The PTS of the previously filtered video frame. It's NAN if undefined.
15240 @item prev_selected_pts
15241 The PTS of the last previously filtered video frame. It's NAN if undefined.
15243 @item prev_selected_t
15244 The PTS of the last previously selected video frame. It's NAN if undefined.
15247 The PTS of the first video frame in the video. It's NAN if undefined.
15250 The time of the first video frame in the video. It's NAN if undefined.
15252 @item pict_type @emph{(video only)}
15253 The type of the filtered frame. It can assume one of the following
15265 @item interlace_type @emph{(video only)}
15266 The frame interlace type. It can assume one of the following values:
15269 The frame is progressive (not interlaced).
15271 The frame is top-field-first.
15273 The frame is bottom-field-first.
15276 @item consumed_sample_n @emph{(audio only)}
15277 the number of selected samples before the current frame
15279 @item samples_n @emph{(audio only)}
15280 the number of samples in the current frame
15282 @item sample_rate @emph{(audio only)}
15283 the input sample rate
15286 This is 1 if the filtered frame is a key-frame, 0 otherwise.
15289 the position in the file of the filtered frame, -1 if the information
15290 is not available (e.g. for synthetic video)
15292 @item scene @emph{(video only)}
15293 value between 0 and 1 to indicate a new scene; a low value reflects a low
15294 probability for the current frame to introduce a new scene, while a higher
15295 value means the current frame is more likely to be one (see the example below)
15297 @item concatdec_select
15298 The concat demuxer can select only part of a concat input file by setting an
15299 inpoint and an outpoint, but the output packets may not be entirely contained
15300 in the selected interval. By using this variable, it is possible to skip frames
15301 generated by the concat demuxer which are not exactly contained in the selected
15304 This works by comparing the frame pts against the @var{lavf.concat.start_time}
15305 and the @var{lavf.concat.duration} packet metadata values which are also
15306 present in the decoded frames.
15308 The @var{concatdec_select} variable is -1 if the frame pts is at least
15309 start_time and either the duration metadata is missing or the frame pts is less
15310 than start_time + duration, 0 otherwise, and NaN if the start_time metadata is
15313 That basically means that an input frame is selected if its pts is within the
15314 interval set by the concat demuxer.
15318 The default value of the select expression is "1".
15320 @subsection Examples
15324 Select all frames in input:
15329 The example above is the same as:
15341 Select only I-frames:
15343 select='eq(pict_type\,I)'
15347 Select one frame every 100:
15349 select='not(mod(n\,100))'
15353 Select only frames contained in the 10-20 time interval:
15355 select=between(t\,10\,20)
15359 Select only I frames contained in the 10-20 time interval:
15361 select=between(t\,10\,20)*eq(pict_type\,I)
15365 Select frames with a minimum distance of 10 seconds:
15367 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
15371 Use aselect to select only audio frames with samples number > 100:
15373 aselect='gt(samples_n\,100)'
15377 Create a mosaic of the first scenes:
15379 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
15382 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
15386 Send even and odd frames to separate outputs, and compose them:
15388 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
15392 Select useful frames from an ffconcat file which is using inpoints and
15393 outpoints but where the source files are not intra frame only.
15395 ffmpeg -copyts -vsync 0 -segment_time_metadata 1 -i input.ffconcat -vf select=concatdec_select -af aselect=concatdec_select output.avi
15399 @section sendcmd, asendcmd
15401 Send commands to filters in the filtergraph.
15403 These filters read commands to be sent to other filters in the
15406 @code{sendcmd} must be inserted between two video filters,
15407 @code{asendcmd} must be inserted between two audio filters, but apart
15408 from that they act the same way.
15410 The specification of commands can be provided in the filter arguments
15411 with the @var{commands} option, or in a file specified by the
15412 @var{filename} option.
15414 These filters accept the following options:
15417 Set the commands to be read and sent to the other filters.
15419 Set the filename of the commands to be read and sent to the other
15423 @subsection Commands syntax
15425 A commands description consists of a sequence of interval
15426 specifications, comprising a list of commands to be executed when a
15427 particular event related to that interval occurs. The occurring event
15428 is typically the current frame time entering or leaving a given time
15431 An interval is specified by the following syntax:
15433 @var{START}[-@var{END}] @var{COMMANDS};
15436 The time interval is specified by the @var{START} and @var{END} times.
15437 @var{END} is optional and defaults to the maximum time.
15439 The current frame time is considered within the specified interval if
15440 it is included in the interval [@var{START}, @var{END}), that is when
15441 the time is greater or equal to @var{START} and is lesser than
15444 @var{COMMANDS} consists of a sequence of one or more command
15445 specifications, separated by ",", relating to that interval. The
15446 syntax of a command specification is given by:
15448 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
15451 @var{FLAGS} is optional and specifies the type of events relating to
15452 the time interval which enable sending the specified command, and must
15453 be a non-null sequence of identifier flags separated by "+" or "|" and
15454 enclosed between "[" and "]".
15456 The following flags are recognized:
15459 The command is sent when the current frame timestamp enters the
15460 specified interval. In other words, the command is sent when the
15461 previous frame timestamp was not in the given interval, and the
15465 The command is sent when the current frame timestamp leaves the
15466 specified interval. In other words, the command is sent when the
15467 previous frame timestamp was in the given interval, and the
15471 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
15474 @var{TARGET} specifies the target of the command, usually the name of
15475 the filter class or a specific filter instance name.
15477 @var{COMMAND} specifies the name of the command for the target filter.
15479 @var{ARG} is optional and specifies the optional list of argument for
15480 the given @var{COMMAND}.
15482 Between one interval specification and another, whitespaces, or
15483 sequences of characters starting with @code{#} until the end of line,
15484 are ignored and can be used to annotate comments.
15486 A simplified BNF description of the commands specification syntax
15489 @var{COMMAND_FLAG} ::= "enter" | "leave"
15490 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
15491 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
15492 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
15493 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
15494 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
15497 @subsection Examples
15501 Specify audio tempo change at second 4:
15503 asendcmd=c='4.0 atempo tempo 1.5',atempo
15507 Specify a list of drawtext and hue commands in a file.
15509 # show text in the interval 5-10
15510 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
15511 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
15513 # desaturate the image in the interval 15-20
15514 15.0-20.0 [enter] hue s 0,
15515 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
15517 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
15519 # apply an exponential saturation fade-out effect, starting from time 25
15520 25 [enter] hue s exp(25-t)
15523 A filtergraph allowing to read and process the above command list
15524 stored in a file @file{test.cmd}, can be specified with:
15526 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
15531 @section setpts, asetpts
15533 Change the PTS (presentation timestamp) of the input frames.
15535 @code{setpts} works on video frames, @code{asetpts} on audio frames.
15537 This filter accepts the following options:
15542 The expression which is evaluated for each frame to construct its timestamp.
15546 The expression is evaluated through the eval API and can contain the following
15551 frame rate, only defined for constant frame-rate video
15554 The presentation timestamp in input
15557 The count of the input frame for video or the number of consumed samples,
15558 not including the current frame for audio, starting from 0.
15560 @item NB_CONSUMED_SAMPLES
15561 The number of consumed samples, not including the current frame (only
15564 @item NB_SAMPLES, S
15565 The number of samples in the current frame (only audio)
15567 @item SAMPLE_RATE, SR
15568 The audio sample rate.
15571 The PTS of the first frame.
15574 the time in seconds of the first frame
15577 State whether the current frame is interlaced.
15580 the time in seconds of the current frame
15583 original position in the file of the frame, or undefined if undefined
15584 for the current frame
15587 The previous input PTS.
15590 previous input time in seconds
15593 The previous output PTS.
15596 previous output time in seconds
15599 The wallclock (RTC) time in microseconds. This is deprecated, use time(0)
15603 The wallclock (RTC) time at the start of the movie in microseconds.
15606 The timebase of the input timestamps.
15610 @subsection Examples
15614 Start counting PTS from zero
15616 setpts=PTS-STARTPTS
15620 Apply fast motion effect:
15626 Apply slow motion effect:
15632 Set fixed rate of 25 frames per second:
15638 Set fixed rate 25 fps with some jitter:
15640 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
15644 Apply an offset of 10 seconds to the input PTS:
15650 Generate timestamps from a "live source" and rebase onto the current timebase:
15652 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
15656 Generate timestamps by counting samples:
15663 @section settb, asettb
15665 Set the timebase to use for the output frames timestamps.
15666 It is mainly useful for testing timebase configuration.
15668 It accepts the following parameters:
15673 The expression which is evaluated into the output timebase.
15677 The value for @option{tb} is an arithmetic expression representing a
15678 rational. The expression can contain the constants "AVTB" (the default
15679 timebase), "intb" (the input timebase) and "sr" (the sample rate,
15680 audio only). Default value is "intb".
15682 @subsection Examples
15686 Set the timebase to 1/25:
15692 Set the timebase to 1/10:
15698 Set the timebase to 1001/1000:
15704 Set the timebase to 2*intb:
15710 Set the default timebase value:
15717 Convert input audio to a video output representing frequency spectrum
15718 logarithmically using Brown-Puckette constant Q transform algorithm with
15719 direct frequency domain coefficient calculation (but the transform itself
15720 is not really constant Q, instead the Q factor is actually variable/clamped),
15721 with musical tone scale, from E0 to D#10.
15723 The filter accepts the following options:
15727 Specify the video size for the output. It must be even. For the syntax of this option,
15728 check the @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15729 Default value is @code{1920x1080}.
15732 Set the output frame rate. Default value is @code{25}.
15735 Set the bargraph height. It must be even. Default value is @code{-1} which
15736 computes the bargraph height automatically.
15739 Set the axis height. It must be even. Default value is @code{-1} which computes
15740 the axis height automatically.
15743 Set the sonogram height. It must be even. Default value is @code{-1} which
15744 computes the sonogram height automatically.
15747 Set the fullhd resolution. This option is deprecated, use @var{size}, @var{s}
15748 instead. Default value is @code{1}.
15750 @item sono_v, volume
15751 Specify the sonogram volume expression. It can contain variables:
15754 the @var{bar_v} evaluated expression
15755 @item frequency, freq, f
15756 the frequency where it is evaluated
15757 @item timeclamp, tc
15758 the value of @var{timeclamp} option
15762 @item a_weighting(f)
15763 A-weighting of equal loudness
15764 @item b_weighting(f)
15765 B-weighting of equal loudness
15766 @item c_weighting(f)
15767 C-weighting of equal loudness.
15769 Default value is @code{16}.
15771 @item bar_v, volume2
15772 Specify the bargraph volume expression. It can contain variables:
15775 the @var{sono_v} evaluated expression
15776 @item frequency, freq, f
15777 the frequency where it is evaluated
15778 @item timeclamp, tc
15779 the value of @var{timeclamp} option
15783 @item a_weighting(f)
15784 A-weighting of equal loudness
15785 @item b_weighting(f)
15786 B-weighting of equal loudness
15787 @item c_weighting(f)
15788 C-weighting of equal loudness.
15790 Default value is @code{sono_v}.
15792 @item sono_g, gamma
15793 Specify the sonogram gamma. Lower gamma makes the spectrum more contrast,
15794 higher gamma makes the spectrum having more range. Default value is @code{3}.
15795 Acceptable range is @code{[1, 7]}.
15797 @item bar_g, gamma2
15798 Specify the bargraph gamma. Default value is @code{1}. Acceptable range is
15801 @item timeclamp, tc
15802 Specify the transform timeclamp. At low frequency, there is trade-off between
15803 accuracy in time domain and frequency domain. If timeclamp is lower,
15804 event in time domain is represented more accurately (such as fast bass drum),
15805 otherwise event in frequency domain is represented more accurately
15806 (such as bass guitar). Acceptable range is @code{[0.1, 1]}. Default value is @code{0.17}.
15809 Specify the transform base frequency. Default value is @code{20.01523126408007475},
15810 which is frequency 50 cents below E0. Acceptable range is @code{[10, 100000]}.
15813 Specify the transform end frequency. Default value is @code{20495.59681441799654},
15814 which is frequency 50 cents above D#10. Acceptable range is @code{[10, 100000]}.
15817 This option is deprecated and ignored.
15820 Specify the transform length in time domain. Use this option to control accuracy
15821 trade-off between time domain and frequency domain at every frequency sample.
15822 It can contain variables:
15824 @item frequency, freq, f
15825 the frequency where it is evaluated
15826 @item timeclamp, tc
15827 the value of @var{timeclamp} option.
15829 Default value is @code{384*tc/(384+tc*f)}.
15832 Specify the transform count for every video frame. Default value is @code{6}.
15833 Acceptable range is @code{[1, 30]}.
15836 Specify the transform count for every single pixel. Default value is @code{0},
15837 which makes it computed automatically. Acceptable range is @code{[0, 10]}.
15840 Specify font file for use with freetype to draw the axis. If not specified,
15841 use embedded font. Note that drawing with font file or embedded font is not
15842 implemented with custom @var{basefreq} and @var{endfreq}, use @var{axisfile}
15846 Specify font color expression. This is arithmetic expression that should return
15847 integer value 0xRRGGBB. It can contain variables:
15849 @item frequency, freq, f
15850 the frequency where it is evaluated
15851 @item timeclamp, tc
15852 the value of @var{timeclamp} option
15857 midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
15858 @item r(x), g(x), b(x)
15859 red, green, and blue value of intensity x.
15861 Default value is @code{st(0, (midi(f)-59.5)/12);
15862 st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
15863 r(1-ld(1)) + b(ld(1))}.
15866 Specify image file to draw the axis. This option override @var{fontfile} and
15867 @var{fontcolor} option.
15870 Enable/disable drawing text to the axis. If it is set to @code{0}, drawing to
15871 the axis is disabled, ignoring @var{fontfile} and @var{axisfile} option.
15872 Default value is @code{1}.
15876 @subsection Examples
15880 Playing audio while showing the spectrum:
15882 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
15886 Same as above, but with frame rate 30 fps:
15888 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
15892 Playing at 1280x720:
15894 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=s=1280x720:count=4 [out0]'
15898 Disable sonogram display:
15904 A1 and its harmonics: A1, A2, (near)E3, A3:
15906 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),
15907 asplit[a][out1]; [a] showcqt [out0]'
15911 Same as above, but with more accuracy in frequency domain:
15913 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),
15914 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
15920 bar_v=10:sono_v=bar_v*a_weighting(f)
15924 Custom gamma, now spectrum is linear to the amplitude.
15930 Custom tlength equation:
15932 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)))'
15936 Custom fontcolor and fontfile, C-note is colored green, others are colored blue:
15938 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))':fontfile=myfont.ttf
15942 Custom frequency range with custom axis using image file:
15944 axisfile=myaxis.png:basefreq=40:endfreq=10000
15950 Convert input audio to video output representing the audio power spectrum.
15951 Audio amplitude is on Y-axis while frequency is on X-axis.
15953 The filter accepts the following options:
15957 Specify size of video. For the syntax of this option, check the
15958 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
15959 Default is @code{1024x512}.
15963 This set how each frequency bin will be represented.
15965 It accepts the following values:
15971 Default is @code{bar}.
15974 Set amplitude scale.
15976 It accepts the following values:
15990 Default is @code{log}.
15993 Set frequency scale.
15995 It accepts the following values:
16004 Reverse logarithmic scale.
16006 Default is @code{lin}.
16011 It accepts the following values:
16027 Default is @code{w2048}
16030 Set windowing function.
16032 It accepts the following values:
16050 Default is @code{hanning}.
16053 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
16054 which means optimal overlap for selected window function will be picked.
16057 Set time averaging. Setting this to 0 will display current maximal peaks.
16058 Default is @code{1}, which means time averaging is disabled.
16061 Specify list of colors separated by space or by '|' which will be used to
16062 draw channel frequencies. Unrecognized or missing colors will be replaced
16066 Set channel display mode.
16068 It accepts the following values:
16073 Default is @code{combined}.
16077 @anchor{showspectrum}
16078 @section showspectrum
16080 Convert input audio to a video output, representing the audio frequency
16083 The filter accepts the following options:
16087 Specify the video size for the output. For the syntax of this option, check the
16088 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16089 Default value is @code{640x512}.
16092 Specify how the spectrum should slide along the window.
16094 It accepts the following values:
16097 the samples start again on the left when they reach the right
16099 the samples scroll from right to left
16101 the samples scroll from left to right
16103 frames are only produced when the samples reach the right
16106 Default value is @code{replace}.
16109 Specify display mode.
16111 It accepts the following values:
16114 all channels are displayed in the same row
16116 all channels are displayed in separate rows
16119 Default value is @samp{combined}.
16122 Specify display color mode.
16124 It accepts the following values:
16127 each channel is displayed in a separate color
16129 each channel is displayed using the same color scheme
16131 each channel is displayed using the rainbow color scheme
16133 each channel is displayed using the moreland color scheme
16135 each channel is displayed using the nebulae color scheme
16137 each channel is displayed using the fire color scheme
16139 each channel is displayed using the fiery color scheme
16141 each channel is displayed using the fruit color scheme
16143 each channel is displayed using the cool color scheme
16146 Default value is @samp{channel}.
16149 Specify scale used for calculating intensity color values.
16151 It accepts the following values:
16156 square root, default
16167 Default value is @samp{sqrt}.
16170 Set saturation modifier for displayed colors. Negative values provide
16171 alternative color scheme. @code{0} is no saturation at all.
16172 Saturation must be in [-10.0, 10.0] range.
16173 Default value is @code{1}.
16176 Set window function.
16178 It accepts the following values:
16198 Default value is @code{hann}.
16201 Set orientation of time vs frequency axis. Can be @code{vertical} or
16202 @code{horizontal}. Default is @code{vertical}.
16205 Set ratio of overlap window. Default value is @code{0}.
16206 When value is @code{1} overlap is set to recommended size for specific
16207 window function currently used.
16210 Set scale gain for calculating intensity color values.
16211 Default value is @code{1}.
16214 Set which data to display. Can be @code{magnitude}, default or @code{phase}.
16217 The usage is very similar to the showwaves filter; see the examples in that
16220 @subsection Examples
16224 Large window with logarithmic color scaling:
16226 showspectrum=s=1280x480:scale=log
16230 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
16232 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
16233 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
16237 @section showspectrumpic
16239 Convert input audio to a single video frame, representing the audio frequency
16242 The filter accepts the following options:
16246 Specify the video size for the output. For the syntax of this option, check the
16247 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16248 Default value is @code{4096x2048}.
16251 Specify display mode.
16253 It accepts the following values:
16256 all channels are displayed in the same row
16258 all channels are displayed in separate rows
16260 Default value is @samp{combined}.
16263 Specify display color mode.
16265 It accepts the following values:
16268 each channel is displayed in a separate color
16270 each channel is displayed using the same color scheme
16272 each channel is displayed using the rainbow color scheme
16274 each channel is displayed using the moreland color scheme
16276 each channel is displayed using the nebulae color scheme
16278 each channel is displayed using the fire color scheme
16280 each channel is displayed using the fiery color scheme
16282 each channel is displayed using the fruit color scheme
16284 each channel is displayed using the cool color scheme
16286 Default value is @samp{intensity}.
16289 Specify scale used for calculating intensity color values.
16291 It accepts the following values:
16296 square root, default
16306 Default value is @samp{log}.
16309 Set saturation modifier for displayed colors. Negative values provide
16310 alternative color scheme. @code{0} is no saturation at all.
16311 Saturation must be in [-10.0, 10.0] range.
16312 Default value is @code{1}.
16315 Set window function.
16317 It accepts the following values:
16336 Default value is @code{hann}.
16339 Set orientation of time vs frequency axis. Can be @code{vertical} or
16340 @code{horizontal}. Default is @code{vertical}.
16343 Set scale gain for calculating intensity color values.
16344 Default value is @code{1}.
16347 Draw time and frequency axes and legends. Default is enabled.
16350 @subsection Examples
16354 Extract an audio spectrogram of a whole audio track
16355 in a 1024x1024 picture using @command{ffmpeg}:
16357 ffmpeg -i audio.flac -lavfi showspectrumpic=s=1024x1024 spectrogram.png
16361 @section showvolume
16363 Convert input audio volume to a video output.
16365 The filter accepts the following options:
16372 Set border width, allowed range is [0, 5]. Default is 1.
16375 Set channel width, allowed range is [80, 8192]. Default is 400.
16378 Set channel height, allowed range is [1, 900]. Default is 20.
16381 Set fade, allowed range is [0.001, 1]. Default is 0.95.
16384 Set volume color expression.
16386 The expression can use the following variables:
16390 Current max volume of channel in dB.
16393 Current channel number, starting from 0.
16397 If set, displays channel names. Default is enabled.
16400 If set, displays volume values. Default is enabled.
16403 Set orientation, can be @code{horizontal} or @code{vertical},
16404 default is @code{horizontal}.
16407 Set step size, allowed range s [0, 5]. Default is 0, which means
16413 Convert input audio to a video output, representing the samples waves.
16415 The filter accepts the following options:
16419 Specify the video size for the output. For the syntax of this option, check the
16420 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16421 Default value is @code{600x240}.
16426 Available values are:
16429 Draw a point for each sample.
16432 Draw a vertical line for each sample.
16435 Draw a point for each sample and a line between them.
16438 Draw a centered vertical line for each sample.
16441 Default value is @code{point}.
16444 Set the number of samples which are printed on the same column. A
16445 larger value will decrease the frame rate. Must be a positive
16446 integer. This option can be set only if the value for @var{rate}
16447 is not explicitly specified.
16450 Set the (approximate) output frame rate. This is done by setting the
16451 option @var{n}. Default value is "25".
16453 @item split_channels
16454 Set if channels should be drawn separately or overlap. Default value is 0.
16457 Set colors separated by '|' which are going to be used for drawing of each channel.
16460 Set amplitude scale. Can be linear @code{lin} or logarithmic @code{log}.
16465 @subsection Examples
16469 Output the input file audio and the corresponding video representation
16472 amovie=a.mp3,asplit[out0],showwaves[out1]
16476 Create a synthetic signal and show it with showwaves, forcing a
16477 frame rate of 30 frames per second:
16479 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
16483 @section showwavespic
16485 Convert input audio to a single video frame, representing the samples waves.
16487 The filter accepts the following options:
16491 Specify the video size for the output. For the syntax of this option, check the
16492 @ref{video size syntax,,"Video size" section in the ffmpeg-utils manual,ffmpeg-utils}.
16493 Default value is @code{600x240}.
16495 @item split_channels
16496 Set if channels should be drawn separately or overlap. Default value is 0.
16499 Set colors separated by '|' which are going to be used for drawing of each channel.
16502 Set amplitude scale. Can be linear @code{lin} or logarithmic @code{log}.
16506 @subsection Examples
16510 Extract a channel split representation of the wave form of a whole audio track
16511 in a 1024x800 picture using @command{ffmpeg}:
16513 ffmpeg -i audio.flac -lavfi showwavespic=split_channels=1:s=1024x800 waveform.png
16517 Colorize the waveform with colorchannelmixer. This example will make
16518 the waveform a green color approximately RGB(66,217,150). Additional
16519 channels will be shades of this color.
16521 ffmpeg -i audio.mp3 -filter_complex "showwavespic,colorchannelmixer=rr=66/255:gg=217/255:bb=150/255" waveform.png
16525 @section spectrumsynth
16527 Sythesize audio from 2 input video spectrums, first input stream represents
16528 magnitude across time and second represents phase across time.
16529 The filter will transform from frequency domain as displayed in videos back
16530 to time domain as presented in audio output.
16532 This filter is primarly created for reversing processed @ref{showspectrum}
16533 filter outputs, but can synthesize sound from other spectrograms too.
16534 But in such case results are going to be poor if the phase data is not
16535 available, because in such cases phase data need to be recreated, usually
16536 its just recreated from random noise.
16537 For best results use gray only output (@code{channel} color mode in
16538 @ref{showspectrum} filter) and @code{log} scale for magnitude video and
16539 @code{lin} scale for phase video. To produce phase, for 2nd video, use
16540 @code{data} option. Inputs videos should generally use @code{fullframe}
16541 slide mode as that saves resources needed for decoding video.
16543 The filter accepts the following options:
16547 Specify sample rate of output audio, the sample rate of audio from which
16548 spectrum was generated may differ.
16551 Set number of channels represented in input video spectrums.
16554 Set scale which was used when generating magnitude input spectrum.
16555 Can be @code{lin} or @code{log}. Default is @code{log}.
16558 Set slide which was used when generating inputs spectrums.
16559 Can be @code{replace}, @code{scroll}, @code{fullframe} or @code{rscroll}.
16560 Default is @code{fullframe}.
16563 Set window function used for resynthesis.
16566 Set window overlap. In range @code{[0, 1]}. Default is @code{1},
16567 which means optimal overlap for selected window function will be picked.
16570 Set orientation of input videos. Can be @code{vertical} or @code{horizontal}.
16571 Default is @code{vertical}.
16574 @subsection Examples
16578 First create magnitude and phase videos from audio, assuming audio is stereo with 44100 sample rate,
16579 then resynthesize videos back to audio with spectrumsynth:
16581 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
16582 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
16583 ffmpeg -i magnitude.nut -i phase.nut -lavfi spectrumsynth=channels=2:sample_rate=44100:win_func=hann:overlap=0.875:slide=fullframe output.flac
16587 @section split, asplit
16589 Split input into several identical outputs.
16591 @code{asplit} works with audio input, @code{split} with video.
16593 The filter accepts a single parameter which specifies the number of outputs. If
16594 unspecified, it defaults to 2.
16596 @subsection Examples
16600 Create two separate outputs from the same input:
16602 [in] split [out0][out1]
16606 To create 3 or more outputs, you need to specify the number of
16609 [in] asplit=3 [out0][out1][out2]
16613 Create two separate outputs from the same input, one cropped and
16616 [in] split [splitout1][splitout2];
16617 [splitout1] crop=100:100:0:0 [cropout];
16618 [splitout2] pad=200:200:100:100 [padout];
16622 Create 5 copies of the input audio with @command{ffmpeg}:
16624 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
16630 Receive commands sent through a libzmq client, and forward them to
16631 filters in the filtergraph.
16633 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
16634 must be inserted between two video filters, @code{azmq} between two
16637 To enable these filters you need to install the libzmq library and
16638 headers and configure FFmpeg with @code{--enable-libzmq}.
16640 For more information about libzmq see:
16641 @url{http://www.zeromq.org/}
16643 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
16644 receives messages sent through a network interface defined by the
16645 @option{bind_address} option.
16647 The received message must be in the form:
16649 @var{TARGET} @var{COMMAND} [@var{ARG}]
16652 @var{TARGET} specifies the target of the command, usually the name of
16653 the filter class or a specific filter instance name.
16655 @var{COMMAND} specifies the name of the command for the target filter.
16657 @var{ARG} is optional and specifies the optional argument list for the
16658 given @var{COMMAND}.
16660 Upon reception, the message is processed and the corresponding command
16661 is injected into the filtergraph. Depending on the result, the filter
16662 will send a reply to the client, adopting the format:
16664 @var{ERROR_CODE} @var{ERROR_REASON}
16668 @var{MESSAGE} is optional.
16670 @subsection Examples
16672 Look at @file{tools/zmqsend} for an example of a zmq client which can
16673 be used to send commands processed by these filters.
16675 Consider the following filtergraph generated by @command{ffplay}
16677 ffplay -dumpgraph 1 -f lavfi "
16678 color=s=100x100:c=red [l];
16679 color=s=100x100:c=blue [r];
16680 nullsrc=s=200x100, zmq [bg];
16681 [bg][l] overlay [bg+l];
16682 [bg+l][r] overlay=x=100 "
16685 To change the color of the left side of the video, the following
16686 command can be used:
16688 echo Parsed_color_0 c yellow | tools/zmqsend
16691 To change the right side:
16693 echo Parsed_color_1 c pink | tools/zmqsend
16696 @c man end MULTIMEDIA FILTERS
16698 @chapter Multimedia Sources
16699 @c man begin MULTIMEDIA SOURCES
16701 Below is a description of the currently available multimedia sources.
16705 This is the same as @ref{movie} source, except it selects an audio
16711 Read audio and/or video stream(s) from a movie container.
16713 It accepts the following parameters:
16717 The name of the resource to read (not necessarily a file; it can also be a
16718 device or a stream accessed through some protocol).
16720 @item format_name, f
16721 Specifies the format assumed for the movie to read, and can be either
16722 the name of a container or an input device. If not specified, the
16723 format is guessed from @var{movie_name} or by probing.
16725 @item seek_point, sp
16726 Specifies the seek point in seconds. The frames will be output
16727 starting from this seek point. The parameter is evaluated with
16728 @code{av_strtod}, so the numerical value may be suffixed by an IS
16729 postfix. The default value is "0".
16732 Specifies the streams to read. Several streams can be specified,
16733 separated by "+". The source will then have as many outputs, in the
16734 same order. The syntax is explained in the ``Stream specifiers''
16735 section in the ffmpeg manual. Two special names, "dv" and "da" specify
16736 respectively the default (best suited) video and audio stream. Default
16737 is "dv", or "da" if the filter is called as "amovie".
16739 @item stream_index, si
16740 Specifies the index of the video stream to read. If the value is -1,
16741 the most suitable video stream will be automatically selected. The default
16742 value is "-1". Deprecated. If the filter is called "amovie", it will select
16743 audio instead of video.
16746 Specifies how many times to read the stream in sequence.
16747 If the value is less than 1, the stream will be read again and again.
16748 Default value is "1".
16750 Note that when the movie is looped the source timestamps are not
16751 changed, so it will generate non monotonically increasing timestamps.
16754 It allows overlaying a second video on top of the main input of
16755 a filtergraph, as shown in this graph:
16757 input -----------> deltapts0 --> overlay --> output
16760 movie --> scale--> deltapts1 -------+
16762 @subsection Examples
16766 Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
16767 on top of the input labelled "in":
16769 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
16770 [in] setpts=PTS-STARTPTS [main];
16771 [main][over] overlay=16:16 [out]
16775 Read from a video4linux2 device, and overlay it on top of the input
16778 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
16779 [in] setpts=PTS-STARTPTS [main];
16780 [main][over] overlay=16:16 [out]
16784 Read the first video stream and the audio stream with id 0x81 from
16785 dvd.vob; the video is connected to the pad named "video" and the audio is
16786 connected to the pad named "audio":
16788 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
16792 @c man end MULTIMEDIA SOURCES