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
118 recognized by the @option{-filter}/@option{-vf} and @option{-filter_complex}
119 options in @command{ffmpeg} and @option{-vf} in @command{ffplay}, and by the
120 @code{avfilter_graph_parse()}/@code{avfilter_graph_parse2()} functions defined in
121 @file{libavfilter/avfilter.h}.
123 A filterchain consists of a sequence of connected filters, each one
124 connected to the previous one in the sequence. A filterchain is
125 represented by a list of ","-separated filter descriptions.
127 A filtergraph consists of a sequence of filterchains. A sequence of
128 filterchains is represented by a list of ";"-separated filterchain
131 A filter is represented by a string of the form:
132 [@var{in_link_1}]...[@var{in_link_N}]@var{filter_name}=@var{arguments}[@var{out_link_1}]...[@var{out_link_M}]
134 @var{filter_name} is the name of the filter class of which the
135 described filter is an instance of, and has to be the name of one of
136 the filter classes registered in the program.
137 The name of the filter class is optionally followed by a string
140 @var{arguments} is a string which contains the parameters used to
141 initialize the filter instance. It may have one of two forms:
145 A ':'-separated list of @var{key=value} pairs.
148 A ':'-separated list of @var{value}. In this case, the keys are assumed to be
149 the option names in the order they are declared. E.g. the @code{fade} filter
150 declares three options in this order -- @option{type}, @option{start_frame} and
151 @option{nb_frames}. Then the parameter list @var{in:0:30} means that the value
152 @var{in} is assigned to the option @option{type}, @var{0} to
153 @option{start_frame} and @var{30} to @option{nb_frames}.
156 A ':'-separated list of mixed direct @var{value} and long @var{key=value}
157 pairs. The direct @var{value} must precede the @var{key=value} pairs, and
158 follow the same constraints order of the previous point. The following
159 @var{key=value} pairs can be set in any preferred order.
163 If the option value itself is a list of items (e.g. the @code{format} filter
164 takes a list of pixel formats), the items in the list are usually separated by
167 The list of arguments can be quoted using the character "'" as initial
168 and ending mark, and the character '\' for escaping the characters
169 within the quoted text; otherwise the argument string is considered
170 terminated when the next special character (belonging to the set
171 "[]=;,") is encountered.
173 The name and arguments of the filter are optionally preceded and
174 followed by a list of link labels.
175 A link label allows one to name a link and associate it to a filter output
176 or input pad. The preceding labels @var{in_link_1}
177 ... @var{in_link_N}, are associated to the filter input pads,
178 the following labels @var{out_link_1} ... @var{out_link_M}, are
179 associated to the output pads.
181 When two link labels with the same name are found in the
182 filtergraph, a link between the corresponding input and output pad is
185 If an output pad is not labelled, it is linked by default to the first
186 unlabelled input pad of the next filter in the filterchain.
187 For example in the filterchain
189 nullsrc, split[L1], [L2]overlay, nullsink
191 the split filter instance has two output pads, and the overlay filter
192 instance two input pads. The first output pad of split is labelled
193 "L1", the first input pad of overlay is labelled "L2", and the second
194 output pad of split is linked to the second input pad of overlay,
195 which are both unlabelled.
197 In a complete filterchain all the unlabelled filter input and output
198 pads must be connected. A filtergraph is considered valid if all the
199 filter input and output pads of all the filterchains are connected.
201 Libavfilter will automatically insert @ref{scale} filters where format
202 conversion is required. It is possible to specify swscale flags
203 for those automatically inserted scalers by prepending
204 @code{sws_flags=@var{flags};}
205 to the filtergraph description.
207 Here is a BNF description of the filtergraph syntax:
209 @var{NAME} ::= sequence of alphanumeric characters and '_'
210 @var{LINKLABEL} ::= "[" @var{NAME} "]"
211 @var{LINKLABELS} ::= @var{LINKLABEL} [@var{LINKLABELS}]
212 @var{FILTER_ARGUMENTS} ::= sequence of chars (possibly quoted)
213 @var{FILTER} ::= [@var{LINKLABELS}] @var{NAME} ["=" @var{FILTER_ARGUMENTS}] [@var{LINKLABELS}]
214 @var{FILTERCHAIN} ::= @var{FILTER} [,@var{FILTERCHAIN}]
215 @var{FILTERGRAPH} ::= [sws_flags=@var{flags};] @var{FILTERCHAIN} [;@var{FILTERGRAPH}]
218 @section Notes on filtergraph escaping
220 Filtergraph description composition entails several levels of
221 escaping. See @ref{quoting_and_escaping,,the "Quoting and escaping"
222 section in the ffmpeg-utils(1) manual,ffmpeg-utils} for more
223 information about the employed escaping procedure.
225 A first level escaping affects the content of each filter option
226 value, which may contain the special character @code{:} used to
227 separate values, or one of the escaping characters @code{\'}.
229 A second level escaping affects the whole filter description, which
230 may contain the escaping characters @code{\'} or the special
231 characters @code{[],;} used by the filtergraph description.
233 Finally, when you specify a filtergraph on a shell commandline, you
234 need to perform a third level escaping for the shell special
235 characters contained within it.
237 For example, consider the following string to be embedded in
238 the @ref{drawtext} filter description @option{text} value:
240 this is a 'string': may contain one, or more, special characters
243 This string contains the @code{'} special escaping character, and the
244 @code{:} special character, so it needs to be escaped in this way:
246 text=this is a \'string\'\: may contain one, or more, special characters
249 A second level of escaping is required when embedding the filter
250 description in a filtergraph description, in order to escape all the
251 filtergraph special characters. Thus the example above becomes:
253 drawtext=text=this is a \\\'string\\\'\\: may contain one\, or more\, special characters
255 (note that in addition to the @code{\'} escaping special characters,
256 also @code{,} needs to be escaped).
258 Finally an additional level of escaping is needed when writing the
259 filtergraph description in a shell command, which depends on the
260 escaping rules of the adopted shell. For example, assuming that
261 @code{\} is special and needs to be escaped with another @code{\}, the
262 previous string will finally result in:
264 -vf "drawtext=text=this is a \\\\\\'string\\\\\\'\\\\: may contain one\\, or more\\, special characters"
267 @chapter Timeline editing
269 Some filters support a generic @option{enable} option. For the filters
270 supporting timeline editing, this option can be set to an expression which is
271 evaluated before sending a frame to the filter. If the evaluation is non-zero,
272 the filter will be enabled, otherwise the frame will be sent unchanged to the
273 next filter in the filtergraph.
275 The expression accepts the following values:
278 timestamp expressed in seconds, NAN if the input timestamp is unknown
281 sequential number of the input frame, starting from 0
284 the position in the file of the input frame, NAN if unknown
287 Additionally, these filters support an @option{enable} command that can be used
288 to re-define the expression.
290 Like any other filtering option, the @option{enable} option follows the same
293 For example, to enable a blur filter (@ref{smartblur}) from 10 seconds to 3
294 minutes, and a @ref{curves} filter starting at 3 seconds:
296 smartblur = enable='between(t,10,3*60)',
297 curves = enable='gte(t,3)' : preset=cross_process
300 @c man end FILTERGRAPH DESCRIPTION
302 @chapter Audio Filters
303 @c man begin AUDIO FILTERS
305 When you configure your FFmpeg build, you can disable any of the
306 existing filters using @code{--disable-filters}.
307 The configure output will show the audio filters included in your
310 Below is a description of the currently available audio filters.
314 Convert the input audio format to the specified formats.
316 @emph{This filter is deprecated. Use @ref{aformat} instead.}
318 The filter accepts a string of the form:
319 "@var{sample_format}:@var{channel_layout}".
321 @var{sample_format} specifies the sample format, and can be a string or the
322 corresponding numeric value defined in @file{libavutil/samplefmt.h}. Use 'p'
323 suffix for a planar sample format.
325 @var{channel_layout} specifies the channel layout, and can be a string
326 or the corresponding number value defined in @file{libavutil/channel_layout.h}.
328 The special parameter "auto", signifies that the filter will
329 automatically select the output format depending on the output filter.
335 Convert input to float, planar, stereo:
341 Convert input to unsigned 8-bit, automatically select out channel layout:
349 Delay one or more audio channels.
351 Samples in delayed channel are filled with silence.
353 The filter accepts the following option:
357 Set list of delays in milliseconds for each channel separated by '|'.
358 At least one delay greater than 0 should be provided.
359 Unused delays will be silently ignored. If number of given delays is
360 smaller than number of channels all remaining channels will not be delayed.
367 Delay first channel by 1.5 seconds, the third channel by 0.5 seconds and leave
368 the second channel (and any other channels that may be present) unchanged.
376 Apply echoing to the input audio.
378 Echoes are reflected sound and can occur naturally amongst mountains
379 (and sometimes large buildings) when talking or shouting; digital echo
380 effects emulate this behaviour and are often used to help fill out the
381 sound of a single instrument or vocal. The time difference between the
382 original signal and the reflection is the @code{delay}, and the
383 loudness of the reflected signal is the @code{decay}.
384 Multiple echoes can have different delays and decays.
386 A description of the accepted parameters follows.
390 Set input gain of reflected signal. Default is @code{0.6}.
393 Set output gain of reflected signal. Default is @code{0.3}.
396 Set list of time intervals in milliseconds between original signal and reflections
397 separated by '|'. Allowed range for each @code{delay} is @code{(0 - 90000.0]}.
398 Default is @code{1000}.
401 Set list of loudnesses of reflected signals separated by '|'.
402 Allowed range for each @code{decay} is @code{(0 - 1.0]}.
403 Default is @code{0.5}.
410 Make it sound as if there are twice as many instruments as are actually playing:
412 aecho=0.8:0.88:60:0.4
416 If delay is very short, then it sound like a (metallic) robot playing music:
422 A longer delay will sound like an open air concert in the mountains:
424 aecho=0.8:0.9:1000:0.3
428 Same as above but with one more mountain:
430 aecho=0.8:0.9:1000|1800:0.3|0.25
436 Modify an audio signal according to the specified expressions.
438 This filter accepts one or more expressions (one for each channel),
439 which are evaluated and used to modify a corresponding audio signal.
441 It accepts the following parameters:
445 Set the '|'-separated expressions list for each separate channel. If
446 the number of input channels is greater than the number of
447 expressions, the last specified expression is used for the remaining
450 @item channel_layout, c
451 Set output channel layout. If not specified, the channel layout is
452 specified by the number of expressions. If set to @samp{same}, it will
453 use by default the same input channel layout.
456 Each expression in @var{exprs} can contain the following constants and functions:
460 channel number of the current expression
463 number of the evaluated sample, starting from 0
469 time of the evaluated sample expressed in seconds
472 @item nb_out_channels
473 input and output number of channels
476 the value of input channel with number @var{CH}
479 Note: this filter is slow. For faster processing you should use a
488 aeval=val(ch)/2:c=same
492 Invert phase of the second channel:
500 Apply fade-in/out effect to input audio.
502 A description of the accepted parameters follows.
506 Specify the effect type, can be either @code{in} for fade-in, or
507 @code{out} for a fade-out effect. Default is @code{in}.
509 @item start_sample, ss
510 Specify the number of the start sample for starting to apply the fade
511 effect. Default is 0.
514 Specify the number of samples for which the fade effect has to last. At
515 the end of the fade-in effect the output audio will have the same
516 volume as the input audio, at the end of the fade-out transition
517 the output audio will be silence. Default is 44100.
520 Specify the start time of the fade effect. Default is 0.
521 The value must be specified as a time duration; see
522 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
523 for the accepted syntax.
524 If set this option is used instead of @var{start_sample}.
527 Specify the duration of the fade effect. See
528 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
529 for the accepted syntax.
530 At the end of the fade-in effect the output audio will have the same
531 volume as the input audio, at the end of the fade-out transition
532 the output audio will be silence.
533 By default the duration is determined by @var{nb_samples}.
534 If set this option is used instead of @var{nb_samples}.
537 Set curve for fade transition.
539 It accepts the following values:
542 select triangular, linear slope (default)
544 select quarter of sine wave
546 select half of sine wave
548 select exponential sine wave
552 select inverted parabola
568 Fade in first 15 seconds of audio:
574 Fade out last 25 seconds of a 900 seconds audio:
576 afade=t=out:st=875:d=25
583 Set output format constraints for the input audio. The framework will
584 negotiate the most appropriate format to minimize conversions.
586 It accepts the following parameters:
590 A '|'-separated list of requested sample formats.
593 A '|'-separated list of requested sample rates.
595 @item channel_layouts
596 A '|'-separated list of requested channel layouts.
598 See @ref{channel layout syntax,,the Channel Layout section in the ffmpeg-utils(1) manual,ffmpeg-utils}
599 for the required syntax.
602 If a parameter is omitted, all values are allowed.
604 Force the output to either unsigned 8-bit or signed 16-bit stereo
606 aformat=sample_fmts=u8|s16:channel_layouts=stereo
611 Apply a two-pole all-pass filter with central frequency (in Hz)
612 @var{frequency}, and filter-width @var{width}.
613 An all-pass filter changes the audio's frequency to phase relationship
614 without changing its frequency to amplitude relationship.
616 The filter accepts the following options:
623 Set method to specify band-width of filter.
636 Specify the band-width of a filter in width_type units.
641 Merge two or more audio streams into a single multi-channel stream.
643 The filter accepts the following options:
648 Set the number of inputs. Default is 2.
652 If the channel layouts of the inputs are disjoint, and therefore compatible,
653 the channel layout of the output will be set accordingly and the channels
654 will be reordered as necessary. If the channel layouts of the inputs are not
655 disjoint, the output will have all the channels of the first input then all
656 the channels of the second input, in that order, and the channel layout of
657 the output will be the default value corresponding to the total number of
660 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
661 is FC+BL+BR, then the output will be in 5.1, with the channels in the
662 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
663 first input, b1 is the first channel of the second input).
665 On the other hand, if both input are in stereo, the output channels will be
666 in the default order: a1, a2, b1, b2, and the channel layout will be
667 arbitrarily set to 4.0, which may or may not be the expected value.
669 All inputs must have the same sample rate, and format.
671 If inputs do not have the same duration, the output will stop with the
678 Merge two mono files into a stereo stream:
680 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
684 Multiple merges assuming 1 video stream and 6 audio streams in @file{input.mkv}:
686 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
692 Mixes multiple audio inputs into a single output.
694 Note that this filter only supports float samples (the @var{amerge}
695 and @var{pan} audio filters support many formats). If the @var{amix}
696 input has integer samples then @ref{aresample} will be automatically
697 inserted to perform the conversion to float samples.
701 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
703 will mix 3 input audio streams to a single output with the same duration as the
704 first input and a dropout transition time of 3 seconds.
706 It accepts the following parameters:
710 The number of inputs. If unspecified, it defaults to 2.
713 How to determine the end-of-stream.
717 The duration of the longest input. (default)
720 The duration of the shortest input.
723 The duration of the first input.
727 @item dropout_transition
728 The transition time, in seconds, for volume renormalization when an input
729 stream ends. The default value is 2 seconds.
735 Pass the audio source unchanged to the output.
739 Pad the end of an audio stream with silence.
741 This can be used together with @command{ffmpeg} @option{-shortest} to
742 extend audio streams to the same length as the video stream.
744 A description of the accepted options follows.
748 Set silence packet size. Default value is 4096.
751 Set the number of samples of silence to add to the end. After the
752 value is reached, the stream is terminated. This option is mutually
753 exclusive with @option{whole_len}.
756 Set the minimum total number of samples in the output audio stream. If
757 the value is longer than the input audio length, silence is added to
758 the end, until the value is reached. This option is mutually exclusive
759 with @option{pad_len}.
762 If neither the @option{pad_len} nor the @option{whole_len} option is
763 set, the filter will add silence to the end of the input stream
770 Add 1024 samples of silence to the end of the input:
776 Make sure the audio output will contain at least 10000 samples, pad
777 the input with silence if required:
783 Use @command{ffmpeg} to pad the audio input with silence, so that the
784 video stream will always result the shortest and will be converted
785 until the end in the output file when using the @option{shortest}
788 ffmpeg -i VIDEO -i AUDIO -filter_complex "[1:0]apad" -shortest OUTPUT
793 Add a phasing effect to the input audio.
795 A phaser filter creates series of peaks and troughs in the frequency spectrum.
796 The position of the peaks and troughs are modulated so that they vary over time, creating a sweeping effect.
798 A description of the accepted parameters follows.
802 Set input gain. Default is 0.4.
805 Set output gain. Default is 0.74
808 Set delay in milliseconds. Default is 3.0.
811 Set decay. Default is 0.4.
814 Set modulation speed in Hz. Default is 0.5.
817 Set modulation type. Default is triangular.
819 It accepts the following values:
829 Resample the input audio to the specified parameters, using the
830 libswresample library. If none are specified then the filter will
831 automatically convert between its input and output.
833 This filter is also able to stretch/squeeze the audio data to make it match
834 the timestamps or to inject silence / cut out audio to make it match the
835 timestamps, do a combination of both or do neither.
837 The filter accepts the syntax
838 [@var{sample_rate}:]@var{resampler_options}, where @var{sample_rate}
839 expresses a sample rate and @var{resampler_options} is a list of
840 @var{key}=@var{value} pairs, separated by ":". See the
841 ffmpeg-resampler manual for the complete list of supported options.
847 Resample the input audio to 44100Hz:
853 Stretch/squeeze samples to the given timestamps, with a maximum of 1000
854 samples per second compensation:
860 @section asetnsamples
862 Set the number of samples per each output audio frame.
864 The last output packet may contain a different number of samples, as
865 the filter will flush all the remaining samples when the input audio
868 The filter accepts the following options:
872 @item nb_out_samples, n
873 Set the number of frames per each output audio frame. The number is
874 intended as the number of samples @emph{per each channel}.
875 Default value is 1024.
878 If set to 1, the filter will pad the last audio frame with zeroes, so
879 that the last frame will contain the same number of samples as the
880 previous ones. Default value is 1.
883 For example, to set the number of per-frame samples to 1234 and
884 disable padding for the last frame, use:
886 asetnsamples=n=1234:p=0
891 Set the sample rate without altering the PCM data.
892 This will result in a change of speed and pitch.
894 The filter accepts the following options:
898 Set the output sample rate. Default is 44100 Hz.
903 Show a line containing various information for each input audio frame.
904 The input audio is not modified.
906 The shown line contains a sequence of key/value pairs of the form
907 @var{key}:@var{value}.
909 The following values are shown in the output:
913 The (sequential) number of the input frame, starting from 0.
916 The presentation timestamp of the input frame, in time base units; the time base
917 depends on the filter input pad, and is usually 1/@var{sample_rate}.
920 The presentation timestamp of the input frame in seconds.
923 position of the frame in the input stream, -1 if this information in
924 unavailable and/or meaningless (for example in case of synthetic audio)
933 The sample rate for the audio frame.
936 The number of samples (per channel) in the frame.
939 The Adler-32 checksum (printed in hexadecimal) of the audio data. For planar
940 audio, the data is treated as if all the planes were concatenated.
942 @item plane_checksums
943 A list of Adler-32 checksums for each data plane.
948 Display time domain statistical information about the audio channels.
949 Statistics are calculated and displayed for each audio channel and,
950 where applicable, an overall figure is also given.
952 It accepts the following option:
955 Short window length in seconds, used for peak and trough RMS measurement.
956 Default is @code{0.05} (50 miliseconds). Allowed range is @code{[0.1 - 10]}.
959 A description of each shown parameter follows:
963 Mean amplitude displacement from zero.
966 Minimal sample level.
969 Maximal sample level.
973 Standard peak and RMS level measured in dBFS.
977 Peak and trough values for RMS level measured over a short window.
980 Standard ratio of peak to RMS level (note: not in dB).
983 Flatness (i.e. consecutive samples with the same value) of the signal at its peak levels
984 (i.e. either @var{Min level} or @var{Max level}).
987 Number of occasions (not the number of samples) that the signal attained either
988 @var{Min level} or @var{Max level}.
993 Forward two audio streams and control the order the buffers are forwarded.
995 The filter accepts the following options:
999 Set the expression deciding which stream should be
1000 forwarded next: if the result is negative, the first stream is forwarded; if
1001 the result is positive or zero, the second stream is forwarded. It can use
1002 the following variables:
1006 number of buffers forwarded so far on each stream
1008 number of samples forwarded so far on each stream
1010 current timestamp of each stream
1013 The default value is @code{t1-t2}, which means to always forward the stream
1014 that has a smaller timestamp.
1017 @subsection Examples
1019 Stress-test @code{amerge} by randomly sending buffers on the wrong
1020 input, while avoiding too much of a desynchronization:
1022 amovie=file.ogg [a] ; amovie=file.mp3 [b] ;
1023 [a] [b] astreamsync=(2*random(1))-1+tanh(5*(t1-t2)) [a2] [b2] ;
1029 Synchronize audio data with timestamps by squeezing/stretching it and/or
1030 dropping samples/adding silence when needed.
1032 This filter is not built by default, please use @ref{aresample} to do squeezing/stretching.
1034 It accepts the following parameters:
1038 Enable stretching/squeezing the data to make it match the timestamps. Disabled
1039 by default. When disabled, time gaps are covered with silence.
1042 The minimum difference between timestamps and audio data (in seconds) to trigger
1043 adding/dropping samples. The default value is 0.1. If you get an imperfect
1044 sync with this filter, try setting this parameter to 0.
1047 The maximum compensation in samples per second. Only relevant with compensate=1.
1048 The default value is 500.
1051 Assume that the first PTS should be this value. The time base is 1 / sample
1052 rate. This allows for padding/trimming at the start of the stream. By default,
1053 no assumption is made about the first frame's expected PTS, so no padding or
1054 trimming is done. For example, this could be set to 0 to pad the beginning with
1055 silence if an audio stream starts after the video stream or to trim any samples
1056 with a negative PTS due to encoder delay.
1064 The filter accepts exactly one parameter, the audio tempo. If not
1065 specified then the filter will assume nominal 1.0 tempo. Tempo must
1066 be in the [0.5, 2.0] range.
1068 @subsection Examples
1072 Slow down audio to 80% tempo:
1078 To speed up audio to 125% tempo:
1086 Trim the input so that the output contains one continuous subpart of the input.
1088 It accepts the following parameters:
1091 Timestamp (in seconds) of the start of the section to keep. I.e. the audio
1092 sample with the timestamp @var{start} will be the first sample in the output.
1095 Specify time of the first audio sample that will be dropped, i.e. the
1096 audio sample immediately preceding the one with the timestamp @var{end} will be
1097 the last sample in the output.
1100 Same as @var{start}, except this option sets the start timestamp in samples
1104 Same as @var{end}, except this option sets the end timestamp in samples instead
1108 The maximum duration of the output in seconds.
1111 The number of the first sample that should be output.
1114 The number of the first sample that should be dropped.
1117 @option{start}, @option{end}, and @option{duration} are expressed as time
1118 duration specifications; see
1119 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}.
1121 Note that the first two sets of the start/end options and the @option{duration}
1122 option look at the frame timestamp, while the _sample options simply count the
1123 samples that pass through the filter. So start/end_pts and start/end_sample will
1124 give different results when the timestamps are wrong, inexact or do not start at
1125 zero. Also note that this filter does not modify the timestamps. If you wish
1126 to have the output timestamps start at zero, insert the asetpts filter after the
1129 If multiple start or end options are set, this filter tries to be greedy and
1130 keep all samples that match at least one of the specified constraints. To keep
1131 only the part that matches all the constraints at once, chain multiple atrim
1134 The defaults are such that all the input is kept. So it is possible to set e.g.
1135 just the end values to keep everything before the specified time.
1140 Drop everything except the second minute of input:
1142 ffmpeg -i INPUT -af atrim=60:120
1146 Keep only the first 1000 samples:
1148 ffmpeg -i INPUT -af atrim=end_sample=1000
1155 Apply a two-pole Butterworth band-pass filter with central
1156 frequency @var{frequency}, and (3dB-point) band-width width.
1157 The @var{csg} option selects a constant skirt gain (peak gain = Q)
1158 instead of the default: constant 0dB peak gain.
1159 The filter roll off at 6dB per octave (20dB per decade).
1161 The filter accepts the following options:
1165 Set the filter's central frequency. Default is @code{3000}.
1168 Constant skirt gain if set to 1. Defaults to 0.
1171 Set method to specify band-width of filter.
1184 Specify the band-width of a filter in width_type units.
1189 Apply a two-pole Butterworth band-reject filter with central
1190 frequency @var{frequency}, and (3dB-point) band-width @var{width}.
1191 The filter roll off at 6dB per octave (20dB per decade).
1193 The filter accepts the following options:
1197 Set the filter's central frequency. Default is @code{3000}.
1200 Set method to specify band-width of filter.
1213 Specify the band-width of a filter in width_type units.
1218 Boost or cut the bass (lower) frequencies of the audio using a two-pole
1219 shelving filter with a response similar to that of a standard
1220 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1222 The filter accepts the following options:
1226 Give the gain at 0 Hz. Its useful range is about -20
1227 (for a large cut) to +20 (for a large boost).
1228 Beware of clipping when using a positive gain.
1231 Set the filter's central frequency and so can be used
1232 to extend or reduce the frequency range to be boosted or cut.
1233 The default value is @code{100} Hz.
1236 Set method to specify band-width of filter.
1249 Determine how steep is the filter's shelf transition.
1254 Apply a biquad IIR filter with the given coefficients.
1255 Where @var{b0}, @var{b1}, @var{b2} and @var{a0}, @var{a1}, @var{a2}
1256 are the numerator and denominator coefficients respectively.
1259 Bauer stereo to binaural transformation, which improves headphone listening of
1260 stereo audio records.
1262 It accepts the following parameters:
1266 Pre-defined crossfeed level.
1270 Default level (fcut=700, feed=50).
1273 Chu Moy circuit (fcut=700, feed=60).
1276 Jan Meier circuit (fcut=650, feed=95).
1281 Cut frequency (in Hz).
1290 Remap input channels to new locations.
1292 It accepts the following parameters:
1294 @item channel_layout
1295 The channel layout of the output stream.
1298 Map channels from input to output. The argument is a '|'-separated list of
1299 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
1300 @var{in_channel} form. @var{in_channel} can be either the name of the input
1301 channel (e.g. FL for front left) or its index in the input channel layout.
1302 @var{out_channel} is the name of the output channel or its index in the output
1303 channel layout. If @var{out_channel} is not given then it is implicitly an
1304 index, starting with zero and increasing by one for each mapping.
1307 If no mapping is present, the filter will implicitly map input channels to
1308 output channels, preserving indices.
1310 For example, assuming a 5.1+downmix input MOV file,
1312 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL|DR-FR' out.wav
1314 will create an output WAV file tagged as stereo from the downmix channels of
1317 To fix a 5.1 WAV improperly encoded in AAC's native channel order
1319 ffmpeg -i in.wav -filter 'channelmap=1|2|0|5|3|4:channel_layout=5.1' out.wav
1322 @section channelsplit
1324 Split each channel from an input audio stream into a separate output stream.
1326 It accepts the following parameters:
1328 @item channel_layout
1329 The channel layout of the input stream. The default is "stereo".
1332 For example, assuming a stereo input MP3 file,
1334 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
1336 will create an output Matroska file with two audio streams, one containing only
1337 the left channel and the other the right channel.
1339 Split a 5.1 WAV file into per-channel files:
1341 ffmpeg -i in.wav -filter_complex
1342 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
1343 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
1344 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
1349 Compress or expand the audio's dynamic range.
1351 It accepts the following parameters:
1357 A list of times in seconds for each channel over which the instantaneous level
1358 of the input signal is averaged to determine its volume. @var{attacks} refers to
1359 increase of volume and @var{decays} refers to decrease of volume. For most
1360 situations, the attack time (response to the audio getting louder) should be
1361 shorter than the decay time, because the human ear is more sensitive to sudden
1362 loud audio than sudden soft audio. A typical value for attack is 0.3 seconds and
1363 a typical value for decay is 0.8 seconds.
1366 A list of points for the transfer function, specified in dB relative to the
1367 maximum possible signal amplitude. Each key points list must be defined using
1368 the following syntax: @code{x0/y0|x1/y1|x2/y2|....} or
1369 @code{x0/y0 x1/y1 x2/y2 ....}
1371 The input values must be in strictly increasing order but the transfer function
1372 does not have to be monotonically rising. The point @code{0/0} is assumed but
1373 may be overridden (by @code{0/out-dBn}). Typical values for the transfer
1374 function are @code{-70/-70|-60/-20}.
1377 Set the curve radius in dB for all joints. It defaults to 0.01.
1380 Set the additional gain in dB to be applied at all points on the transfer
1381 function. This allows for easy adjustment of the overall gain.
1385 Set an initial volume, in dB, to be assumed for each channel when filtering
1386 starts. This permits the user to supply a nominal level initially, so that, for
1387 example, a very large gain is not applied to initial signal levels before the
1388 companding has begun to operate. A typical value for audio which is initially
1389 quiet is -90 dB. It defaults to 0.
1392 Set a delay, in seconds. The input audio is analyzed immediately, but audio is
1393 delayed before being fed to the volume adjuster. Specifying a delay
1394 approximately equal to the attack/decay times allows the filter to effectively
1395 operate in predictive rather than reactive mode. It defaults to 0.
1399 @subsection Examples
1403 Make music with both quiet and loud passages suitable for listening to in a
1406 compand=.3|.3:1|1:-90/-60|-60/-40|-40/-30|-20/-20:6:0:-90:0.2
1410 A noise gate for when the noise is at a lower level than the signal:
1412 compand=.1|.1:.2|.2:-900/-900|-50.1/-900|-50/-50:.01:0:-90:.1
1416 Here is another noise gate, this time for when the noise is at a higher level
1417 than the signal (making it, in some ways, similar to squelch):
1419 compand=.1|.1:.1|.1:-45.1/-45.1|-45/-900|0/-900:.01:45:-90:.1
1425 Make audio easier to listen to on headphones.
1427 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
1428 so that when listened to on headphones the stereo image is moved from
1429 inside your head (standard for headphones) to outside and in front of
1430 the listener (standard for speakers).
1436 Apply a two-pole peaking equalisation (EQ) filter. With this
1437 filter, the signal-level at and around a selected frequency can
1438 be increased or decreased, whilst (unlike bandpass and bandreject
1439 filters) that at all other frequencies is unchanged.
1441 In order to produce complex equalisation curves, this filter can
1442 be given several times, each with a different central frequency.
1444 The filter accepts the following options:
1448 Set the filter's central frequency in Hz.
1451 Set method to specify band-width of filter.
1464 Specify the band-width of a filter in width_type units.
1467 Set the required gain or attenuation in dB.
1468 Beware of clipping when using a positive gain.
1471 @subsection Examples
1474 Attenuate 10 dB at 1000 Hz, with a bandwidth of 200 Hz:
1476 equalizer=f=1000:width_type=h:width=200:g=-10
1480 Apply 2 dB gain at 1000 Hz with Q 1 and attenuate 5 dB at 100 Hz with Q 2:
1482 equalizer=f=1000:width_type=q:width=1:g=2,equalizer=f=100:width_type=q:width=2:g=-5
1487 Apply a flanging effect to the audio.
1489 The filter accepts the following options:
1493 Set base delay in milliseconds. Range from 0 to 30. Default value is 0.
1496 Set added swep delay in milliseconds. Range from 0 to 10. Default value is 2.
1499 Set percentage regeneneration (delayed signal feedback). Range from -95 to 95.
1503 Set percentage of delayed signal mixed with original. Range from 0 to 100.
1507 Set sweeps per second (Hz). Range from 0.1 to 10. Default value is 0.5.
1510 Set swept wave shape, can be @var{triangular} or @var{sinusoidal}.
1511 Default value is @var{sinusoidal}.
1514 Set swept wave percentage-shift for multi channel. Range from 0 to 100.
1515 Default value is 25.
1518 Set delay-line interpolation, @var{linear} or @var{quadratic}.
1519 Default is @var{linear}.
1524 Apply a high-pass filter with 3dB point frequency.
1525 The filter can be either single-pole, or double-pole (the default).
1526 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
1528 The filter accepts the following options:
1532 Set frequency in Hz. Default is 3000.
1535 Set number of poles. Default is 2.
1538 Set method to specify band-width of filter.
1551 Specify the band-width of a filter in width_type units.
1552 Applies only to double-pole filter.
1553 The default is 0.707q and gives a Butterworth response.
1558 Join multiple input streams into one multi-channel stream.
1560 It accepts the following parameters:
1564 The number of input streams. It defaults to 2.
1566 @item channel_layout
1567 The desired output channel layout. It defaults to stereo.
1570 Map channels from inputs to output. The argument is a '|'-separated list of
1571 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
1572 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
1573 can be either the name of the input channel (e.g. FL for front left) or its
1574 index in the specified input stream. @var{out_channel} is the name of the output
1578 The filter will attempt to guess the mappings when they are not specified
1579 explicitly. It does so by first trying to find an unused matching input channel
1580 and if that fails it picks the first unused input channel.
1582 Join 3 inputs (with properly set channel layouts):
1584 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
1587 Build a 5.1 output from 6 single-channel streams:
1589 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
1590 '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'
1596 Load a LADSPA (Linux Audio Developer's Simple Plugin API) plugin.
1598 To enable compilation of this filter you need to configure FFmpeg with
1599 @code{--enable-ladspa}.
1603 Specifies the name of LADSPA plugin library to load. If the environment
1604 variable @env{LADSPA_PATH} is defined, the LADSPA plugin is searched in
1605 each one of the directories specified by the colon separated list in
1606 @env{LADSPA_PATH}, otherwise in the standard LADSPA paths, which are in
1607 this order: @file{HOME/.ladspa/lib/}, @file{/usr/local/lib/ladspa/},
1608 @file{/usr/lib/ladspa/}.
1611 Specifies the plugin within the library. Some libraries contain only
1612 one plugin, but others contain many of them. If this is not set filter
1613 will list all available plugins within the specified library.
1616 Set the '|' separated list of controls which are zero or more floating point
1617 values that determine the behavior of the loaded plugin (for example delay,
1619 Controls need to be defined using the following syntax:
1620 c0=@var{value0}|c1=@var{value1}|c2=@var{value2}|..., where
1621 @var{valuei} is the value set on the @var{i}-th control.
1622 If @option{controls} is set to @code{help}, all available controls and
1623 their valid ranges are printed.
1625 @item sample_rate, s
1626 Specify the sample rate, default to 44100. Only used if plugin have
1630 Set the number of samples per channel per each output frame, default
1631 is 1024. Only used if plugin have zero inputs.
1634 Set the minimum duration of the sourced audio. See
1635 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
1636 for the accepted syntax.
1637 Note that the resulting duration may be greater than the specified duration,
1638 as the generated audio is always cut at the end of a complete frame.
1639 If not specified, or the expressed duration is negative, the audio is
1640 supposed to be generated forever.
1641 Only used if plugin have zero inputs.
1645 @subsection Examples
1649 List all available plugins within amp (LADSPA example plugin) library:
1655 List all available controls and their valid ranges for @code{vcf_notch}
1656 plugin from @code{VCF} library:
1658 ladspa=f=vcf:p=vcf_notch:c=help
1662 Simulate low quality audio equipment using @code{Computer Music Toolkit} (CMT)
1665 ladspa=file=cmt:plugin=lofi:controls=c0=22|c1=12|c2=12
1669 Add reverberation to the audio using TAP-plugins
1670 (Tom's Audio Processing plugins):
1672 ladspa=file=tap_reverb:tap_reverb
1676 Generate white noise, with 0.2 amplitude:
1678 ladspa=file=cmt:noise_source_white:c=c0=.2
1682 Generate 20 bpm clicks using plugin @code{C* Click - Metronome} from the
1683 @code{C* Audio Plugin Suite} (CAPS) library:
1685 ladspa=file=caps:Click:c=c1=20'
1689 Apply @code{C* Eq10X2 - Stereo 10-band equaliser} effect:
1691 ladspa=caps:Eq10X2:c=c0=-48|c9=-24|c3=12|c4=2
1695 @subsection Commands
1697 This filter supports the following commands:
1700 Modify the @var{N}-th control value.
1702 If the specified value is not valid, it is ignored and prior one is kept.
1707 Apply a low-pass filter with 3dB point frequency.
1708 The filter can be either single-pole or double-pole (the default).
1709 The filter roll off at 6dB per pole per octave (20dB per pole per decade).
1711 The filter accepts the following options:
1715 Set frequency in Hz. Default is 500.
1718 Set number of poles. Default is 2.
1721 Set method to specify band-width of filter.
1734 Specify the band-width of a filter in width_type units.
1735 Applies only to double-pole filter.
1736 The default is 0.707q and gives a Butterworth response.
1741 Mix channels with specific gain levels. The filter accepts the output
1742 channel layout followed by a set of channels definitions.
1744 This filter is also designed to remap efficiently the channels of an audio
1747 The filter accepts parameters of the form:
1748 "@var{l}:@var{outdef}:@var{outdef}:..."
1752 output channel layout or number of channels
1755 output channel specification, of the form:
1756 "@var{out_name}=[@var{gain}*]@var{in_name}[+[@var{gain}*]@var{in_name}...]"
1759 output channel to define, either a channel name (FL, FR, etc.) or a channel
1760 number (c0, c1, etc.)
1763 multiplicative coefficient for the channel, 1 leaving the volume unchanged
1766 input channel to use, see out_name for details; it is not possible to mix
1767 named and numbered input channels
1770 If the `=' in a channel specification is replaced by `<', then the gains for
1771 that specification will be renormalized so that the total is 1, thus
1772 avoiding clipping noise.
1774 @subsection Mixing examples
1776 For example, if you want to down-mix from stereo to mono, but with a bigger
1777 factor for the left channel:
1779 pan=1:c0=0.9*c0+0.1*c1
1782 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
1783 7-channels surround:
1785 pan=stereo: FL < FL + 0.5*FC + 0.6*BL + 0.6*SL : FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
1788 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
1789 that should be preferred (see "-ac" option) unless you have very specific
1792 @subsection Remapping examples
1794 The channel remapping will be effective if, and only if:
1797 @item gain coefficients are zeroes or ones,
1798 @item only one input per channel output,
1801 If all these conditions are satisfied, the filter will notify the user ("Pure
1802 channel mapping detected"), and use an optimized and lossless method to do the
1805 For example, if you have a 5.1 source and want a stereo audio stream by
1806 dropping the extra channels:
1808 pan="stereo: c0=FL : c1=FR"
1811 Given the same source, you can also switch front left and front right channels
1812 and keep the input channel layout:
1814 pan="5.1: c0=c1 : c1=c0 : c2=c2 : c3=c3 : c4=c4 : c5=c5"
1817 If the input is a stereo audio stream, you can mute the front left channel (and
1818 still keep the stereo channel layout) with:
1823 Still with a stereo audio stream input, you can copy the right channel in both
1824 front left and right:
1826 pan="stereo: c0=FR : c1=FR"
1831 ReplayGain scanner filter. This filter takes an audio stream as an input and
1832 outputs it unchanged.
1833 At end of filtering it displays @code{track_gain} and @code{track_peak}.
1837 Convert the audio sample format, sample rate and channel layout. It is
1838 not meant to be used directly.
1840 @section silencedetect
1842 Detect silence in an audio stream.
1844 This filter logs a message when it detects that the input audio volume is less
1845 or equal to a noise tolerance value for a duration greater or equal to the
1846 minimum detected noise duration.
1848 The printed times and duration are expressed in seconds.
1850 The filter accepts the following options:
1854 Set silence duration until notification (default is 2 seconds).
1857 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
1858 specified value) or amplitude ratio. Default is -60dB, or 0.001.
1861 @subsection Examples
1865 Detect 5 seconds of silence with -50dB noise tolerance:
1867 silencedetect=n=-50dB:d=5
1871 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
1872 tolerance in @file{silence.mp3}:
1874 ffmpeg -i silence.mp3 -af silencedetect=noise=0.0001 -f null -
1878 @section silenceremove
1880 Remove silence from the beginning, middle or end of the audio.
1882 The filter accepts the following options:
1886 This value is used to indicate if audio should be trimmed at beginning of
1887 the audio. A value of zero indicates no silence should be trimmed from the
1888 beginning. When specifying a non-zero value, it trims audio up until it
1889 finds non-silence. Normally, when trimming silence from beginning of audio
1890 the @var{start_periods} will be @code{1} but it can be increased to higher
1891 values to trim all audio up to specific count of non-silence periods.
1892 Default value is @code{0}.
1894 @item start_duration
1895 Specify the amount of time that non-silence must be detected before it stops
1896 trimming audio. By increasing the duration, bursts of noises can be treated
1897 as silence and trimmed off. Default value is @code{0}.
1899 @item start_threshold
1900 This indicates what sample value should be treated as silence. For digital
1901 audio, a value of @code{0} may be fine but for audio recorded from analog,
1902 you may wish to increase the value to account for background noise.
1903 Can be specified in dB (in case "dB" is appended to the specified value)
1904 or amplitude ratio. Default value is @code{0}.
1907 Set the count for trimming silence from the end of audio.
1908 To remove silence from the middle of a file, specify a @var{stop_periods}
1909 that is negative. This value is then threated as a positive value and is
1910 used to indicate the effect should restart processing as specified by
1911 @var{start_periods}, making it suitable for removing periods of silence
1912 in the middle of the audio.
1913 Default value is @code{0}.
1916 Specify a duration of silence that must exist before audio is not copied any
1917 more. By specifying a higher duration, silence that is wanted can be left in
1919 Default value is @code{0}.
1921 @item stop_threshold
1922 This is the same as @option{start_threshold} but for trimming silence from
1924 Can be specified in dB (in case "dB" is appended to the specified value)
1925 or amplitude ratio. Default value is @code{0}.
1928 This indicate that @var{stop_duration} length of audio should be left intact
1929 at the beginning of each period of silence.
1930 For example, if you want to remove long pauses between words but do not want
1931 to remove the pauses completely. Default value is @code{0}.
1935 @subsection Examples
1939 The following example shows how this filter can be used to start a recording
1940 that does not contain the delay at the start which usually occurs between
1941 pressing the record button and the start of the performance:
1943 silenceremove=1:5:0.02
1949 Boost or cut treble (upper) frequencies of the audio using a two-pole
1950 shelving filter with a response similar to that of a standard
1951 hi-fi's tone-controls. This is also known as shelving equalisation (EQ).
1953 The filter accepts the following options:
1957 Give the gain at whichever is the lower of ~22 kHz and the
1958 Nyquist frequency. Its useful range is about -20 (for a large cut)
1959 to +20 (for a large boost). Beware of clipping when using a positive gain.
1962 Set the filter's central frequency and so can be used
1963 to extend or reduce the frequency range to be boosted or cut.
1964 The default value is @code{3000} Hz.
1967 Set method to specify band-width of filter.
1980 Determine how steep is the filter's shelf transition.
1985 Adjust the input audio volume.
1987 It accepts the following parameters:
1991 Set audio volume expression.
1993 Output values are clipped to the maximum value.
1995 The output audio volume is given by the relation:
1997 @var{output_volume} = @var{volume} * @var{input_volume}
2000 The default value for @var{volume} is "1.0".
2003 This parameter represents the mathematical precision.
2005 It determines which input sample formats will be allowed, which affects the
2006 precision of the volume scaling.
2010 8-bit fixed-point; this limits input sample format to U8, S16, and S32.
2012 32-bit floating-point; this limits input sample format to FLT. (default)
2014 64-bit floating-point; this limits input sample format to DBL.
2018 Choose the behaviour on encountering ReplayGain side data in input frames.
2022 Remove ReplayGain side data, ignoring its contents (the default).
2025 Ignore ReplayGain side data, but leave it in the frame.
2028 Prefer the track gain, if present.
2031 Prefer the album gain, if present.
2034 @item replaygain_preamp
2035 Pre-amplification gain in dB to apply to the selected replaygain gain.
2037 Default value for @var{replaygain_preamp} is 0.0.
2040 Set when the volume expression is evaluated.
2042 It accepts the following values:
2045 only evaluate expression once during the filter initialization, or
2046 when the @samp{volume} command is sent
2049 evaluate expression for each incoming frame
2052 Default value is @samp{once}.
2055 The volume expression can contain the following parameters.
2059 frame number (starting at zero)
2062 @item nb_consumed_samples
2063 number of samples consumed by the filter
2065 number of samples in the current frame
2067 original frame position in the file
2073 PTS at start of stream
2075 time at start of stream
2081 last set volume value
2084 Note that when @option{eval} is set to @samp{once} only the
2085 @var{sample_rate} and @var{tb} variables are available, all other
2086 variables will evaluate to NAN.
2088 @subsection Commands
2090 This filter supports the following commands:
2093 Modify the volume expression.
2094 The command accepts the same syntax of the corresponding option.
2096 If the specified expression is not valid, it is kept at its current
2098 @item replaygain_noclip
2099 Prevent clipping by limiting the gain applied.
2101 Default value for @var{replaygain_noclip} is 1.
2105 @subsection Examples
2109 Halve the input audio volume:
2113 volume=volume=-6.0206dB
2116 In all the above example the named key for @option{volume} can be
2117 omitted, for example like in:
2123 Increase input audio power by 6 decibels using fixed-point precision:
2125 volume=volume=6dB:precision=fixed
2129 Fade volume after time 10 with an annihilation period of 5 seconds:
2131 volume='if(lt(t,10),1,max(1-(t-10)/5,0))':eval=frame
2135 @section volumedetect
2137 Detect the volume of the input video.
2139 The filter has no parameters. The input is not modified. Statistics about
2140 the volume will be printed in the log when the input stream end is reached.
2142 In particular it will show the mean volume (root mean square), maximum
2143 volume (on a per-sample basis), and the beginning of a histogram of the
2144 registered volume values (from the maximum value to a cumulated 1/1000 of
2147 All volumes are in decibels relative to the maximum PCM value.
2149 @subsection Examples
2151 Here is an excerpt of the output:
2153 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
2154 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
2155 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
2156 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
2157 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
2158 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
2159 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
2160 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
2161 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
2167 The mean square energy is approximately -27 dB, or 10^-2.7.
2169 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
2171 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
2174 In other words, raising the volume by +4 dB does not cause any clipping,
2175 raising it by +5 dB causes clipping for 6 samples, etc.
2177 @c man end AUDIO FILTERS
2179 @chapter Audio Sources
2180 @c man begin AUDIO SOURCES
2182 Below is a description of the currently available audio sources.
2186 Buffer audio frames, and make them available to the filter chain.
2188 This source is mainly intended for a programmatic use, in particular
2189 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
2191 It accepts the following parameters:
2195 The timebase which will be used for timestamps of submitted frames. It must be
2196 either a floating-point number or in @var{numerator}/@var{denominator} form.
2199 The sample rate of the incoming audio buffers.
2202 The sample format of the incoming audio buffers.
2203 Either a sample format name or its corresponging integer representation from
2204 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
2206 @item channel_layout
2207 The channel layout of the incoming audio buffers.
2208 Either a channel layout name from channel_layout_map in
2209 @file{libavutil/channel_layout.c} or its corresponding integer representation
2210 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
2213 The number of channels of the incoming audio buffers.
2214 If both @var{channels} and @var{channel_layout} are specified, then they
2219 @subsection Examples
2222 abuffer=sample_rate=44100:sample_fmt=s16p:channel_layout=stereo
2225 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
2226 Since the sample format with name "s16p" corresponds to the number
2227 6 and the "stereo" channel layout corresponds to the value 0x3, this is
2230 abuffer=sample_rate=44100:sample_fmt=6:channel_layout=0x3
2235 Generate an audio signal specified by an expression.
2237 This source accepts in input one or more expressions (one for each
2238 channel), which are evaluated and used to generate a corresponding
2241 This source accepts the following options:
2245 Set the '|'-separated expressions list for each separate channel. In case the
2246 @option{channel_layout} option is not specified, the selected channel layout
2247 depends on the number of provided expressions. Otherwise the last
2248 specified expression is applied to the remaining output channels.
2250 @item channel_layout, c
2251 Set the channel layout. The number of channels in the specified layout
2252 must be equal to the number of specified expressions.
2255 Set the minimum duration of the sourced audio. See
2256 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
2257 for the accepted syntax.
2258 Note that the resulting duration may be greater than the specified
2259 duration, as the generated audio is always cut at the end of a
2262 If not specified, or the expressed duration is negative, the audio is
2263 supposed to be generated forever.
2266 Set the number of samples per channel per each output frame,
2269 @item sample_rate, s
2270 Specify the sample rate, default to 44100.
2273 Each expression in @var{exprs} can contain the following constants:
2277 number of the evaluated sample, starting from 0
2280 time of the evaluated sample expressed in seconds, starting from 0
2287 @subsection Examples
2297 Generate a sin signal with frequency of 440 Hz, set sample rate to
2300 aevalsrc="sin(440*2*PI*t):s=8000"
2304 Generate a two channels signal, specify the channel layout (Front
2305 Center + Back Center) explicitly:
2307 aevalsrc="sin(420*2*PI*t)|cos(430*2*PI*t):c=FC|BC"
2311 Generate white noise:
2313 aevalsrc="-2+random(0)"
2317 Generate an amplitude modulated signal:
2319 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
2323 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
2325 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) | 0.1*sin(2*PI*(360+2.5/2)*t)"
2332 The null audio source, return unprocessed audio frames. It is mainly useful
2333 as a template and to be employed in analysis / debugging tools, or as
2334 the source for filters which ignore the input data (for example the sox
2337 This source accepts the following options:
2341 @item channel_layout, cl
2343 Specifies the channel layout, and can be either an integer or a string
2344 representing a channel layout. The default value of @var{channel_layout}
2347 Check the channel_layout_map definition in
2348 @file{libavutil/channel_layout.c} for the mapping between strings and
2349 channel layout values.
2351 @item sample_rate, r
2352 Specifies the sample rate, and defaults to 44100.
2355 Set the number of samples per requested frames.
2359 @subsection Examples
2363 Set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
2365 anullsrc=r=48000:cl=4
2369 Do the same operation with a more obvious syntax:
2371 anullsrc=r=48000:cl=mono
2375 All the parameters need to be explicitly defined.
2379 Synthesize a voice utterance using the libflite library.
2381 To enable compilation of this filter you need to configure FFmpeg with
2382 @code{--enable-libflite}.
2384 Note that the flite library is not thread-safe.
2386 The filter accepts the following options:
2391 If set to 1, list the names of the available voices and exit
2392 immediately. Default value is 0.
2395 Set the maximum number of samples per frame. Default value is 512.
2398 Set the filename containing the text to speak.
2401 Set the text to speak.
2404 Set the voice to use for the speech synthesis. Default value is
2405 @code{kal}. See also the @var{list_voices} option.
2408 @subsection Examples
2412 Read from file @file{speech.txt}, and synthetize the text using the
2413 standard flite voice:
2415 flite=textfile=speech.txt
2419 Read the specified text selecting the @code{slt} voice:
2421 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
2425 Input text to ffmpeg:
2427 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
2431 Make @file{ffplay} speak the specified text, using @code{flite} and
2432 the @code{lavfi} device:
2434 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
2438 For more information about libflite, check:
2439 @url{http://www.speech.cs.cmu.edu/flite/}
2443 Generate an audio signal made of a sine wave with amplitude 1/8.
2445 The audio signal is bit-exact.
2447 The filter accepts the following options:
2452 Set the carrier frequency. Default is 440 Hz.
2454 @item beep_factor, b
2455 Enable a periodic beep every second with frequency @var{beep_factor} times
2456 the carrier frequency. Default is 0, meaning the beep is disabled.
2458 @item sample_rate, r
2459 Specify the sample rate, default is 44100.
2462 Specify the duration of the generated audio stream.
2464 @item samples_per_frame
2465 Set the number of samples per output frame, default is 1024.
2468 @subsection Examples
2473 Generate a simple 440 Hz sine wave:
2479 Generate a 220 Hz sine wave with a 880 Hz beep each second, for 5 seconds:
2483 sine=frequency=220:beep_factor=4:duration=5
2488 @c man end AUDIO SOURCES
2490 @chapter Audio Sinks
2491 @c man begin AUDIO SINKS
2493 Below is a description of the currently available audio sinks.
2495 @section abuffersink
2497 Buffer audio frames, and make them available to the end of filter chain.
2499 This sink is mainly intended for programmatic use, in particular
2500 through the interface defined in @file{libavfilter/buffersink.h}
2501 or the options system.
2503 It accepts a pointer to an AVABufferSinkContext structure, which
2504 defines the incoming buffers' formats, to be passed as the opaque
2505 parameter to @code{avfilter_init_filter} for initialization.
2508 Null audio sink; do absolutely nothing with the input audio. It is
2509 mainly useful as a template and for use in analysis / debugging
2512 @c man end AUDIO SINKS
2514 @chapter Video Filters
2515 @c man begin VIDEO FILTERS
2517 When you configure your FFmpeg build, you can disable any of the
2518 existing filters using @code{--disable-filters}.
2519 The configure output will show the video filters included in your
2522 Below is a description of the currently available video filters.
2524 @section alphaextract
2526 Extract the alpha component from the input as a grayscale video. This
2527 is especially useful with the @var{alphamerge} filter.
2531 Add or replace the alpha component of the primary input with the
2532 grayscale value of a second input. This is intended for use with
2533 @var{alphaextract} to allow the transmission or storage of frame
2534 sequences that have alpha in a format that doesn't support an alpha
2537 For example, to reconstruct full frames from a normal YUV-encoded video
2538 and a separate video created with @var{alphaextract}, you might use:
2540 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
2543 Since this filter is designed for reconstruction, it operates on frame
2544 sequences without considering timestamps, and terminates when either
2545 input reaches end of stream. This will cause problems if your encoding
2546 pipeline drops frames. If you're trying to apply an image as an
2547 overlay to a video stream, consider the @var{overlay} filter instead.
2551 Same as the @ref{subtitles} filter, except that it doesn't require libavcodec
2552 and libavformat to work. On the other hand, it is limited to ASS (Advanced
2553 Substation Alpha) subtitles files.
2557 Compute the bounding box for the non-black pixels in the input frame
2560 This filter computes the bounding box containing all the pixels with a
2561 luminance value greater than the minimum allowed value.
2562 The parameters describing the bounding box are printed on the filter
2565 The filter accepts the following option:
2569 Set the minimal luminance value. Default is @code{16}.
2572 @section blackdetect
2574 Detect video intervals that are (almost) completely black. Can be
2575 useful to detect chapter transitions, commercials, or invalid
2576 recordings. Output lines contains the time for the start, end and
2577 duration of the detected black interval expressed in seconds.
2579 In order to display the output lines, you need to set the loglevel at
2580 least to the AV_LOG_INFO value.
2582 The filter accepts the following options:
2585 @item black_min_duration, d
2586 Set the minimum detected black duration expressed in seconds. It must
2587 be a non-negative floating point number.
2589 Default value is 2.0.
2591 @item picture_black_ratio_th, pic_th
2592 Set the threshold for considering a picture "black".
2593 Express the minimum value for the ratio:
2595 @var{nb_black_pixels} / @var{nb_pixels}
2598 for which a picture is considered black.
2599 Default value is 0.98.
2601 @item pixel_black_th, pix_th
2602 Set the threshold for considering a pixel "black".
2604 The threshold expresses the maximum pixel luminance value for which a
2605 pixel is considered "black". The provided value is scaled according to
2606 the following equation:
2608 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
2611 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
2612 the input video format, the range is [0-255] for YUV full-range
2613 formats and [16-235] for YUV non full-range formats.
2615 Default value is 0.10.
2618 The following example sets the maximum pixel threshold to the minimum
2619 value, and detects only black intervals of 2 or more seconds:
2621 blackdetect=d=2:pix_th=0.00
2626 Detect frames that are (almost) completely black. Can be useful to
2627 detect chapter transitions or commercials. Output lines consist of
2628 the frame number of the detected frame, the percentage of blackness,
2629 the position in the file if known or -1 and the timestamp in seconds.
2631 In order to display the output lines, you need to set the loglevel at
2632 least to the AV_LOG_INFO value.
2634 It accepts the following parameters:
2639 The percentage of the pixels that have to be below the threshold; it defaults to
2642 @item threshold, thresh
2643 The threshold below which a pixel value is considered black; it defaults to
2650 Blend two video frames into each other.
2652 It takes two input streams and outputs one stream, the first input is the
2653 "top" layer and second input is "bottom" layer.
2654 Output terminates when shortest input terminates.
2656 A description of the accepted options follows.
2664 Set blend mode for specific pixel component or all pixel components in case
2665 of @var{all_mode}. Default value is @code{normal}.
2667 Available values for component modes are:
2700 Set blend opacity for specific pixel component or all pixel components in case
2701 of @var{all_opacity}. Only used in combination with pixel component blend modes.
2708 Set blend expression for specific pixel component or all pixel components in case
2709 of @var{all_expr}. Note that related mode options will be ignored if those are set.
2711 The expressions can use the following variables:
2715 The sequential number of the filtered frame, starting from @code{0}.
2719 the coordinates of the current sample
2723 the width and height of currently filtered plane
2727 Width and height scale depending on the currently filtered plane. It is the
2728 ratio between the corresponding luma plane number of pixels and the current
2729 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
2730 @code{0.5,0.5} for chroma planes.
2733 Time of the current frame, expressed in seconds.
2736 Value of pixel component at current location for first video frame (top layer).
2739 Value of pixel component at current location for second video frame (bottom layer).
2743 Force termination when the shortest input terminates. Default is @code{0}.
2745 Continue applying the last bottom frame after the end of the stream. A value of
2746 @code{0} disable the filter after the last frame of the bottom layer is reached.
2747 Default is @code{1}.
2750 @subsection Examples
2754 Apply transition from bottom layer to top layer in first 10 seconds:
2756 blend=all_expr='A*(if(gte(T,10),1,T/10))+B*(1-(if(gte(T,10),1,T/10)))'
2760 Apply 1x1 checkerboard effect:
2762 blend=all_expr='if(eq(mod(X,2),mod(Y,2)),A,B)'
2766 Apply uncover left effect:
2768 blend=all_expr='if(gte(N*SW+X,W),A,B)'
2772 Apply uncover down effect:
2774 blend=all_expr='if(gte(Y-N*SH,0),A,B)'
2778 Apply uncover up-left effect:
2780 blend=all_expr='if(gte(T*SH*40+Y,H)*gte((T*40*SW+X)*W/H,W),A,B)'
2786 Apply a boxblur algorithm to the input video.
2788 It accepts the following parameters:
2792 @item luma_radius, lr
2793 @item luma_power, lp
2794 @item chroma_radius, cr
2795 @item chroma_power, cp
2796 @item alpha_radius, ar
2797 @item alpha_power, ap
2801 A description of the accepted options follows.
2804 @item luma_radius, lr
2805 @item chroma_radius, cr
2806 @item alpha_radius, ar
2807 Set an expression for the box radius in pixels used for blurring the
2808 corresponding input plane.
2810 The radius value must be a non-negative number, and must not be
2811 greater than the value of the expression @code{min(w,h)/2} for the
2812 luma and alpha planes, and of @code{min(cw,ch)/2} for the chroma
2815 Default value for @option{luma_radius} is "2". If not specified,
2816 @option{chroma_radius} and @option{alpha_radius} default to the
2817 corresponding value set for @option{luma_radius}.
2819 The expressions can contain the following constants:
2823 The input width and height in pixels.
2827 The input chroma image width and height in pixels.
2831 The horizontal and vertical chroma subsample values. For example, for the
2832 pixel format "yuv422p", @var{hsub} is 2 and @var{vsub} is 1.
2835 @item luma_power, lp
2836 @item chroma_power, cp
2837 @item alpha_power, ap
2838 Specify how many times the boxblur filter is applied to the
2839 corresponding plane.
2841 Default value for @option{luma_power} is 2. If not specified,
2842 @option{chroma_power} and @option{alpha_power} default to the
2843 corresponding value set for @option{luma_power}.
2845 A value of 0 will disable the effect.
2848 @subsection Examples
2852 Apply a boxblur filter with the luma, chroma, and alpha radii
2855 boxblur=luma_radius=2:luma_power=1
2860 Set the luma radius to 2, and alpha and chroma radius to 0:
2862 boxblur=2:1:cr=0:ar=0
2866 Set the luma and chroma radii to a fraction of the video dimension:
2868 boxblur=luma_radius=min(h\,w)/10:luma_power=1:chroma_radius=min(cw\,ch)/10:chroma_power=1
2874 Visualize information exported by some codecs.
2876 Some codecs can export information through frames using side-data or other
2877 means. For example, some MPEG based codecs export motion vectors through the
2878 @var{export_mvs} flag in the codec @option{flags2} option.
2880 The filter accepts the following option:
2884 Set motion vectors to visualize.
2886 Available flags for @var{mv} are:
2890 forward predicted MVs of P-frames
2892 forward predicted MVs of B-frames
2894 backward predicted MVs of B-frames
2898 @subsection Examples
2902 Visualizes multi-directionals MVs from P and B-Frames using @command{ffplay}:
2904 ffplay -flags2 +export_mvs input.mpg -vf codecview=mv=pf+bf+bb
2908 @section colorbalance
2909 Modify intensity of primary colors (red, green and blue) of input frames.
2911 The filter allows an input frame to be adjusted in the shadows, midtones or highlights
2912 regions for the red-cyan, green-magenta or blue-yellow balance.
2914 A positive adjustment value shifts the balance towards the primary color, a negative
2915 value towards the complementary color.
2917 The filter accepts the following options:
2923 Adjust red, green and blue shadows (darkest pixels).
2928 Adjust red, green and blue midtones (medium pixels).
2933 Adjust red, green and blue highlights (brightest pixels).
2935 Allowed ranges for options are @code{[-1.0, 1.0]}. Defaults are @code{0}.
2938 @subsection Examples
2942 Add red color cast to shadows:
2948 @section colorchannelmixer
2950 Adjust video input frames by re-mixing color channels.
2952 This filter modifies a color channel by adding the values associated to
2953 the other channels of the same pixels. For example if the value to
2954 modify is red, the output value will be:
2956 @var{red}=@var{red}*@var{rr} + @var{blue}*@var{rb} + @var{green}*@var{rg} + @var{alpha}*@var{ra}
2959 The filter accepts the following options:
2966 Adjust contribution of input red, green, blue and alpha channels for output red channel.
2967 Default is @code{1} for @var{rr}, and @code{0} for @var{rg}, @var{rb} and @var{ra}.
2973 Adjust contribution of input red, green, blue and alpha channels for output green channel.
2974 Default is @code{1} for @var{gg}, and @code{0} for @var{gr}, @var{gb} and @var{ga}.
2980 Adjust contribution of input red, green, blue and alpha channels for output blue channel.
2981 Default is @code{1} for @var{bb}, and @code{0} for @var{br}, @var{bg} and @var{ba}.
2987 Adjust contribution of input red, green, blue and alpha channels for output alpha channel.
2988 Default is @code{1} for @var{aa}, and @code{0} for @var{ar}, @var{ag} and @var{ab}.
2990 Allowed ranges for options are @code{[-2.0, 2.0]}.
2993 @subsection Examples
2997 Convert source to grayscale:
2999 colorchannelmixer=.3:.4:.3:0:.3:.4:.3:0:.3:.4:.3
3002 Simulate sepia tones:
3004 colorchannelmixer=.393:.769:.189:0:.349:.686:.168:0:.272:.534:.131
3008 @section colormatrix
3010 Convert color matrix.
3012 The filter accepts the following options:
3017 Specify the source and destination color matrix. Both values must be
3020 The accepted values are:
3036 For example to convert from BT.601 to SMPTE-240M, use the command:
3038 colormatrix=bt601:smpte240m
3043 Copy the input source unchanged to the output. This is mainly useful for
3048 Crop the input video to given dimensions.
3050 It accepts the following parameters:
3054 The width of the output video. It defaults to @code{iw}.
3055 This expression is evaluated only once during the filter
3059 The height of the output video. It defaults to @code{ih}.
3060 This expression is evaluated only once during the filter
3064 The horizontal position, in the input video, of the left edge of the output
3065 video. It defaults to @code{(in_w-out_w)/2}.
3066 This expression is evaluated per-frame.
3069 The vertical position, in the input video, of the top edge of the output video.
3070 It defaults to @code{(in_h-out_h)/2}.
3071 This expression is evaluated per-frame.
3074 If set to 1 will force the output display aspect ratio
3075 to be the same of the input, by changing the output sample aspect
3076 ratio. It defaults to 0.
3079 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
3080 expressions containing the following constants:
3085 The computed values for @var{x} and @var{y}. They are evaluated for
3090 The input width and height.
3094 These are the same as @var{in_w} and @var{in_h}.
3098 The output (cropped) width and height.
3102 These are the same as @var{out_w} and @var{out_h}.
3105 same as @var{iw} / @var{ih}
3108 input sample aspect ratio
3111 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
3115 horizontal and vertical chroma subsample values. For example for the
3116 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3119 The number of the input frame, starting from 0.
3122 the position in the file of the input frame, NAN if unknown
3125 The timestamp expressed in seconds. It's NAN if the input timestamp is unknown.
3129 The expression for @var{out_w} may depend on the value of @var{out_h},
3130 and the expression for @var{out_h} may depend on @var{out_w}, but they
3131 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
3132 evaluated after @var{out_w} and @var{out_h}.
3134 The @var{x} and @var{y} parameters specify the expressions for the
3135 position of the top-left corner of the output (non-cropped) area. They
3136 are evaluated for each frame. If the evaluated value is not valid, it
3137 is approximated to the nearest valid value.
3139 The expression for @var{x} may depend on @var{y}, and the expression
3140 for @var{y} may depend on @var{x}.
3142 @subsection Examples
3146 Crop area with size 100x100 at position (12,34).
3151 Using named options, the example above becomes:
3153 crop=w=100:h=100:x=12:y=34
3157 Crop the central input area with size 100x100:
3163 Crop the central input area with size 2/3 of the input video:
3165 crop=2/3*in_w:2/3*in_h
3169 Crop the input video central square:
3176 Delimit the rectangle with the top-left corner placed at position
3177 100:100 and the right-bottom corner corresponding to the right-bottom
3178 corner of the input image.
3180 crop=in_w-100:in_h-100:100:100
3184 Crop 10 pixels from the left and right borders, and 20 pixels from
3185 the top and bottom borders
3187 crop=in_w-2*10:in_h-2*20
3191 Keep only the bottom right quarter of the input image:
3193 crop=in_w/2:in_h/2:in_w/2:in_h/2
3197 Crop height for getting Greek harmony:
3199 crop=in_w:1/PHI*in_w
3203 Appply trembling effect:
3205 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)
3209 Apply erratic camera effect depending on timestamp:
3211 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)"
3215 Set x depending on the value of y:
3217 crop=in_w/2:in_h/2:y:10+10*sin(n/10)
3223 Auto-detect the crop size.
3225 It calculates the necessary cropping parameters and prints the
3226 recommended parameters via the logging system. The detected dimensions
3227 correspond to the non-black area of the input video.
3229 It accepts the following parameters:
3234 Set higher black value threshold, which can be optionally specified
3235 from nothing (0) to everything (255). An intensity value greater
3236 to the set value is considered non-black. It defaults to 24.
3239 The value which the width/height should be divisible by. It defaults to
3240 16. The offset is automatically adjusted to center the video. Use 2 to
3241 get only even dimensions (needed for 4:2:2 video). 16 is best when
3242 encoding to most video codecs.
3244 @item reset_count, reset
3245 Set the counter that determines after how many frames cropdetect will
3246 reset the previously detected largest video area and start over to
3247 detect the current optimal crop area. Default value is 0.
3249 This can be useful when channel logos distort the video area. 0
3250 indicates 'never reset', and returns the largest area encountered during
3257 Apply color adjustments using curves.
3259 This filter is similar to the Adobe Photoshop and GIMP curves tools. Each
3260 component (red, green and blue) has its values defined by @var{N} key points
3261 tied from each other using a smooth curve. The x-axis represents the pixel
3262 values from the input frame, and the y-axis the new pixel values to be set for
3265 By default, a component curve is defined by the two points @var{(0;0)} and
3266 @var{(1;1)}. This creates a straight line where each original pixel value is
3267 "adjusted" to its own value, which means no change to the image.
3269 The filter allows you to redefine these two points and add some more. A new
3270 curve (using a natural cubic spline interpolation) will be define to pass
3271 smoothly through all these new coordinates. The new defined points needs to be
3272 strictly increasing over the x-axis, and their @var{x} and @var{y} values must
3273 be in the @var{[0;1]} interval. If the computed curves happened to go outside
3274 the vector spaces, the values will be clipped accordingly.
3276 If there is no key point defined in @code{x=0}, the filter will automatically
3277 insert a @var{(0;0)} point. In the same way, if there is no key point defined
3278 in @code{x=1}, the filter will automatically insert a @var{(1;1)} point.
3280 The filter accepts the following options:
3284 Select one of the available color presets. This option can be used in addition
3285 to the @option{r}, @option{g}, @option{b} parameters; in this case, the later
3286 options takes priority on the preset values.
3287 Available presets are:
3290 @item color_negative
3293 @item increase_contrast
3295 @item linear_contrast
3296 @item medium_contrast
3298 @item strong_contrast
3301 Default is @code{none}.
3303 Set the master key points. These points will define a second pass mapping. It
3304 is sometimes called a "luminance" or "value" mapping. It can be used with
3305 @option{r}, @option{g}, @option{b} or @option{all} since it acts like a
3306 post-processing LUT.
3308 Set the key points for the red component.
3310 Set the key points for the green component.
3312 Set the key points for the blue component.
3314 Set the key points for all components (not including master).
3315 Can be used in addition to the other key points component
3316 options. In this case, the unset component(s) will fallback on this
3317 @option{all} setting.
3319 Specify a Photoshop curves file (@code{.asv}) to import the settings from.
3322 To avoid some filtergraph syntax conflicts, each key points list need to be
3323 defined using the following syntax: @code{x0/y0 x1/y1 x2/y2 ...}.
3325 @subsection Examples
3329 Increase slightly the middle level of blue:
3331 curves=blue='0.5/0.58'
3337 curves=r='0/0.11 .42/.51 1/0.95':g='0.50/0.48':b='0/0.22 .49/.44 1/0.8'
3339 Here we obtain the following coordinates for each components:
3342 @code{(0;0.11) (0.42;0.51) (1;0.95)}
3344 @code{(0;0) (0.50;0.48) (1;1)}
3346 @code{(0;0.22) (0.49;0.44) (1;0.80)}
3350 The previous example can also be achieved with the associated built-in preset:
3352 curves=preset=vintage
3362 Use a Photoshop preset and redefine the points of the green component:
3364 curves=psfile='MyCurvesPresets/purple.asv':green='0.45/0.53'
3370 Denoise frames using 2D DCT (frequency domain filtering).
3372 This filter is not designed for real time.
3374 The filter accepts the following options:
3378 Set the noise sigma constant.
3380 This @var{sigma} defines a hard threshold of @code{3 * sigma}; every DCT
3381 coefficient (absolute value) below this threshold with be dropped.
3383 If you need a more advanced filtering, see @option{expr}.
3385 Default is @code{0}.
3388 Set number overlapping pixels for each block. Since the filter can be slow, you
3389 may want to reduce this value, at the cost of a less effective filter and the
3390 risk of various artefacts.
3392 If the overlapping value doesn't allow to process the whole input width or
3393 height, a warning will be displayed and according borders won't be denoised.
3395 Default value is @var{blocksize}-1, which is the best possible setting.
3398 Set the coefficient factor expression.
3400 For each coefficient of a DCT block, this expression will be evaluated as a
3401 multiplier value for the coefficient.
3403 If this is option is set, the @option{sigma} option will be ignored.
3405 The absolute value of the coefficient can be accessed through the @var{c}
3409 Set the @var{blocksize} using the number of bits. @code{1<<@var{n}} defines the
3410 @var{blocksize}, which is the width and height of the processed blocks.
3412 The default value is @var{3} (8x8) and can be raised to @var{4} for a
3413 @var{blocksize} of 16x16. Note that changing this setting has huge consequences
3414 on the speed processing. Also, a larger block size does not necessarily means a
3418 @subsection Examples
3420 Apply a denoise with a @option{sigma} of @code{4.5}:
3425 The same operation can be achieved using the expression system:
3427 dctdnoiz=e='gte(c, 4.5*3)'
3430 Violent denoise using a block size of @code{16x16}:
3438 Drop duplicated frames at regular intervals.
3440 The filter accepts the following options:
3444 Set the number of frames from which one will be dropped. Setting this to
3445 @var{N} means one frame in every batch of @var{N} frames will be dropped.
3446 Default is @code{5}.
3449 Set the threshold for duplicate detection. If the difference metric for a frame
3450 is less than or equal to this value, then it is declared as duplicate. Default
3454 Set scene change threshold. Default is @code{15}.
3458 Set the size of the x and y-axis blocks used during metric calculations.
3459 Larger blocks give better noise suppression, but also give worse detection of
3460 small movements. Must be a power of two. Default is @code{32}.
3463 Mark main input as a pre-processed input and activate clean source input
3464 stream. This allows the input to be pre-processed with various filters to help
3465 the metrics calculation while keeping the frame selection lossless. When set to
3466 @code{1}, the first stream is for the pre-processed input, and the second
3467 stream is the clean source from where the kept frames are chosen. Default is
3471 Set whether or not chroma is considered in the metric calculations. Default is
3477 Remove judder produced by partially interlaced telecined content.
3479 Judder can be introduced, for instance, by @ref{pullup} filter. If the original
3480 source was partially telecined content then the output of @code{pullup,dejudder}
3481 will have a variable frame rate. May change the recorded frame rate of the
3482 container. Aside from that change, this filter will not affect constant frame
3485 The option available in this filter is:
3489 Specify the length of the window over which the judder repeats.
3491 Accepts any integer greater than 1. Useful values are:
3495 If the original was telecined from 24 to 30 fps (Film to NTSC).
3498 If the original was telecined from 25 to 30 fps (PAL to NTSC).
3501 If a mixture of the two.
3504 The default is @samp{4}.
3509 Suppress a TV station logo by a simple interpolation of the surrounding
3510 pixels. Just set a rectangle covering the logo and watch it disappear
3511 (and sometimes something even uglier appear - your mileage may vary).
3513 It accepts the following parameters:
3518 Specify the top left corner coordinates of the logo. They must be
3523 Specify the width and height of the logo to clear. They must be
3527 Specify the thickness of the fuzzy edge of the rectangle (added to
3528 @var{w} and @var{h}). The default value is 4.
3531 When set to 1, a green rectangle is drawn on the screen to simplify
3532 finding the right @var{x}, @var{y}, @var{w}, and @var{h} parameters.
3533 The default value is 0.
3535 The rectangle is drawn on the outermost pixels which will be (partly)
3536 replaced with interpolated values. The values of the next pixels
3537 immediately outside this rectangle in each direction will be used to
3538 compute the interpolated pixel values inside the rectangle.
3542 @subsection Examples
3546 Set a rectangle covering the area with top left corner coordinates 0,0
3547 and size 100x77, and a band of size 10:
3549 delogo=x=0:y=0:w=100:h=77:band=10
3556 Attempt to fix small changes in horizontal and/or vertical shift. This
3557 filter helps remove camera shake from hand-holding a camera, bumping a
3558 tripod, moving on a vehicle, etc.
3560 The filter accepts the following options:
3568 Specify a rectangular area where to limit the search for motion
3570 If desired the search for motion vectors can be limited to a
3571 rectangular area of the frame defined by its top left corner, width
3572 and height. These parameters have the same meaning as the drawbox
3573 filter which can be used to visualise the position of the bounding
3576 This is useful when simultaneous movement of subjects within the frame
3577 might be confused for camera motion by the motion vector search.
3579 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
3580 then the full frame is used. This allows later options to be set
3581 without specifying the bounding box for the motion vector search.
3583 Default - search the whole frame.
3587 Specify the maximum extent of movement in x and y directions in the
3588 range 0-64 pixels. Default 16.
3591 Specify how to generate pixels to fill blanks at the edge of the
3592 frame. Available values are:
3595 Fill zeroes at blank locations
3597 Original image at blank locations
3599 Extruded edge value at blank locations
3601 Mirrored edge at blank locations
3603 Default value is @samp{mirror}.
3606 Specify the blocksize to use for motion search. Range 4-128 pixels,
3610 Specify the contrast threshold for blocks. Only blocks with more than
3611 the specified contrast (difference between darkest and lightest
3612 pixels) will be considered. Range 1-255, default 125.
3615 Specify the search strategy. Available values are:
3618 Set exhaustive search
3620 Set less exhaustive search.
3622 Default value is @samp{exhaustive}.
3625 If set then a detailed log of the motion search is written to the
3629 If set to 1, specify using OpenCL capabilities, only available if
3630 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
3636 Draw a colored box on the input image.
3638 It accepts the following parameters:
3643 The expressions which specify the top left corner coordinates of the box. It defaults to 0.
3647 The expressions which specify the width and height of the box; if 0 they are interpreted as
3648 the input width and height. It defaults to 0.
3651 Specify the color of the box to write. For the general syntax of this option,
3652 check the "Color" section in the ffmpeg-utils manual. If the special
3653 value @code{invert} is used, the box edge color is the same as the
3654 video with inverted luma.
3657 The expression which sets the thickness of the box edge. Default value is @code{3}.
3659 See below for the list of accepted constants.
3662 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
3663 following constants:
3667 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
3671 horizontal and vertical chroma subsample values. For example for the
3672 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3676 The input width and height.
3679 The input sample aspect ratio.
3683 The x and y offset coordinates where the box is drawn.
3687 The width and height of the drawn box.
3690 The thickness of the drawn box.
3692 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
3693 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
3697 @subsection Examples
3701 Draw a black box around the edge of the input image:
3707 Draw a box with color red and an opacity of 50%:
3709 drawbox=10:20:200:60:red@@0.5
3712 The previous example can be specified as:
3714 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
3718 Fill the box with pink color:
3720 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
3724 Draw a 2-pixel red 2.40:1 mask:
3726 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
3732 Draw a grid on the input image.
3734 It accepts the following parameters:
3739 The expressions which specify the coordinates of some point of grid intersection (meant to configure offset). Both default to 0.
3743 The expressions which specify the width and height of the grid cell, if 0 they are interpreted as the
3744 input width and height, respectively, minus @code{thickness}, so image gets
3745 framed. Default to 0.
3748 Specify the color of the grid. For the general syntax of this option,
3749 check the "Color" section in the ffmpeg-utils manual. If the special
3750 value @code{invert} is used, the grid color is the same as the
3751 video with inverted luma.
3754 The expression which sets the thickness of the grid line. Default value is @code{1}.
3756 See below for the list of accepted constants.
3759 The parameters for @var{x}, @var{y}, @var{w} and @var{h} and @var{t} are expressions containing the
3760 following constants:
3764 The input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}.
3768 horizontal and vertical chroma subsample values. For example for the
3769 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3773 The input grid cell width and height.
3776 The input sample aspect ratio.
3780 The x and y coordinates of some point of grid intersection (meant to configure offset).
3784 The width and height of the drawn cell.
3787 The thickness of the drawn cell.
3789 These constants allow the @var{x}, @var{y}, @var{w}, @var{h} and @var{t} expressions to refer to
3790 each other, so you may for example specify @code{y=x/dar} or @code{h=w/dar}.
3794 @subsection Examples
3798 Draw a grid with cell 100x100 pixels, thickness 2 pixels, with color red and an opacity of 50%:
3800 drawgrid=width=100:height=100:thickness=2:color=red@@0.5
3804 Draw a white 3x3 grid with an opacity of 50%:
3806 drawgrid=w=iw/3:h=ih/3:t=2:c=white@@0.5
3813 Draw a text string or text from a specified file on top of a video, using the
3814 libfreetype library.
3816 To enable compilation of this filter, you need to configure FFmpeg with
3817 @code{--enable-libfreetype}.
3818 To enable default font fallback and the @var{font} option you need to
3819 configure FFmpeg with @code{--enable-libfontconfig}.
3820 To enable the @var{text_shaping} option, you need to configure FFmpeg with
3821 @code{--enable-libfribidi}.
3825 It accepts the following parameters:
3830 Used to draw a box around text using the background color.
3831 The value must be either 1 (enable) or 0 (disable).
3832 The default value of @var{box} is 0.
3835 The color to be used for drawing box around text. For the syntax of this
3836 option, check the "Color" section in the ffmpeg-utils manual.
3838 The default value of @var{boxcolor} is "white".
3841 Set the width of the border to be drawn around the text using @var{bordercolor}.
3842 The default value of @var{borderw} is 0.
3845 Set the color to be used for drawing border around text. For the syntax of this
3846 option, check the "Color" section in the ffmpeg-utils manual.
3848 The default value of @var{bordercolor} is "black".
3851 Select how the @var{text} is expanded. Can be either @code{none},
3852 @code{strftime} (deprecated) or
3853 @code{normal} (default). See the @ref{drawtext_expansion, Text expansion} section
3857 If true, check and fix text coords to avoid clipping.
3860 The color to be used for drawing fonts. For the syntax of this option, check
3861 the "Color" section in the ffmpeg-utils manual.
3863 The default value of @var{fontcolor} is "black".
3865 @item fontcolor_expr
3866 String which is expanded the same way as @var{text} to obtain dynamic
3867 @var{fontcolor} value. By default this option has empty value and is not
3868 processed. When this option is set, it overrides @var{fontcolor} option.
3871 The font family to be used for drawing text. By default Sans.
3874 The font file to be used for drawing text. The path must be included.
3875 This parameter is mandatory if the fontconfig support is disabled.
3878 The font size to be used for drawing text.
3879 The default value of @var{fontsize} is 16.
3882 If set to 1, attempt to shape the text (for example, reverse the order of
3883 right-to-left text and join Arabic characters) before drawing it.
3884 Otherwise, just draw the text exactly as given.
3885 By default 1 (if supported).
3888 The flags to be used for loading the fonts.
3890 The flags map the corresponding flags supported by libfreetype, and are
3891 a combination of the following values:
3898 @item vertical_layout
3899 @item force_autohint
3902 @item ignore_global_advance_width
3904 @item ignore_transform
3910 Default value is "default".
3912 For more information consult the documentation for the FT_LOAD_*
3916 The color to be used for drawing a shadow behind the drawn text. For the
3917 syntax of this option, check the "Color" section in the ffmpeg-utils manual.
3919 The default value of @var{shadowcolor} is "black".
3923 The x and y offsets for the text shadow position with respect to the
3924 position of the text. They can be either positive or negative
3925 values. The default value for both is "0".
3928 The starting frame number for the n/frame_num variable. The default value
3932 The size in number of spaces to use for rendering the tab.
3936 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
3937 format. It can be used with or without text parameter. @var{timecode_rate}
3938 option must be specified.
3940 @item timecode_rate, rate, r
3941 Set the timecode frame rate (timecode only).
3944 The text string to be drawn. The text must be a sequence of UTF-8
3946 This parameter is mandatory if no file is specified with the parameter
3950 A text file containing text to be drawn. The text must be a sequence
3951 of UTF-8 encoded characters.
3953 This parameter is mandatory if no text string is specified with the
3954 parameter @var{text}.
3956 If both @var{text} and @var{textfile} are specified, an error is thrown.
3959 If set to 1, the @var{textfile} will be reloaded before each frame.
3960 Be sure to update it atomically, or it may be read partially, or even fail.
3964 The expressions which specify the offsets where text will be drawn
3965 within the video frame. They are relative to the top/left border of the
3968 The default value of @var{x} and @var{y} is "0".
3970 See below for the list of accepted constants and functions.
3973 The parameters for @var{x} and @var{y} are expressions containing the
3974 following constants and functions:
3978 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
3982 horizontal and vertical chroma subsample values. For example for the
3983 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3986 the height of each text line
3994 @item max_glyph_a, ascent
3995 the maximum distance from the baseline to the highest/upper grid
3996 coordinate used to place a glyph outline point, for all the rendered
3998 It is a positive value, due to the grid's orientation with the Y axis
4001 @item max_glyph_d, descent
4002 the maximum distance from the baseline to the lowest grid coordinate
4003 used to place a glyph outline point, for all the rendered glyphs.
4004 This is a negative value, due to the grid's orientation, with the Y axis
4008 maximum glyph height, that is the maximum height for all the glyphs
4009 contained in the rendered text, it is equivalent to @var{ascent} -
4013 maximum glyph width, that is the maximum width for all the glyphs
4014 contained in the rendered text
4017 the number of input frame, starting from 0
4019 @item rand(min, max)
4020 return a random number included between @var{min} and @var{max}
4023 The input sample aspect ratio.
4026 timestamp expressed in seconds, NAN if the input timestamp is unknown
4029 the height of the rendered text
4032 the width of the rendered text
4036 the x and y offset coordinates where the text is drawn.
4038 These parameters allow the @var{x} and @var{y} expressions to refer
4039 each other, so you can for example specify @code{y=x/dar}.
4042 @anchor{drawtext_expansion}
4043 @subsection Text expansion
4045 If @option{expansion} is set to @code{strftime},
4046 the filter recognizes strftime() sequences in the provided text and
4047 expands them accordingly. Check the documentation of strftime(). This
4048 feature is deprecated.
4050 If @option{expansion} is set to @code{none}, the text is printed verbatim.
4052 If @option{expansion} is set to @code{normal} (which is the default),
4053 the following expansion mechanism is used.
4055 The backslash character '\', followed by any character, always expands to
4056 the second character.
4058 Sequence of the form @code{%@{...@}} are expanded. The text between the
4059 braces is a function name, possibly followed by arguments separated by ':'.
4060 If the arguments contain special characters or delimiters (':' or '@}'),
4061 they should be escaped.
4063 Note that they probably must also be escaped as the value for the
4064 @option{text} option in the filter argument string and as the filter
4065 argument in the filtergraph description, and possibly also for the shell,
4066 that makes up to four levels of escaping; using a text file avoids these
4069 The following functions are available:
4074 The expression evaluation result.
4076 It must take one argument specifying the expression to be evaluated,
4077 which accepts the same constants and functions as the @var{x} and
4078 @var{y} values. Note that not all constants should be used, for
4079 example the text size is not known when evaluating the expression, so
4080 the constants @var{text_w} and @var{text_h} will have an undefined
4083 @item expr_int_format, eif
4084 Evaluate the expression's value and output as formatted integer.
4086 The first argument is the expression to be evaluated, just as for the @var{expr} function.
4087 The second argument specifies the output format. Allowed values are 'x', 'X', 'd' and
4088 'u'. They are treated exactly as in the printf function.
4089 The third parameter is optional and sets the number of positions taken by the output.
4090 It can be used to add padding with zeros from the left.
4093 The time at which the filter is running, expressed in UTC.
4094 It can accept an argument: a strftime() format string.
4097 The time at which the filter is running, expressed in the local time zone.
4098 It can accept an argument: a strftime() format string.
4101 Frame metadata. It must take one argument specifying metadata key.
4104 The frame number, starting from 0.
4107 A 1 character description of the current picture type.
4110 The timestamp of the current frame.
4111 It can take up to two arguments.
4113 The first argument is the format of the timestamp; it defaults to @code{flt}
4114 for seconds as a decimal number with microsecond accuracy; @code{hms} stands
4115 for a formatted @var{[-]HH:MM:SS.mmm} timestamp with millisecond accuracy.
4117 The second argument is an offset added to the timestamp.
4121 @subsection Examples
4125 Draw "Test Text" with font FreeSerif, using the default values for the
4126 optional parameters.
4129 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
4133 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
4134 and y=50 (counting from the top-left corner of the screen), text is
4135 yellow with a red box around it. Both the text and the box have an
4139 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
4140 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
4143 Note that the double quotes are not necessary if spaces are not used
4144 within the parameter list.
4147 Show the text at the center of the video frame:
4149 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h-line_h)/2"
4153 Show a text line sliding from right to left in the last row of the video
4154 frame. The file @file{LONG_LINE} is assumed to contain a single line
4157 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
4161 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
4163 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
4167 Draw a single green letter "g", at the center of the input video.
4168 The glyph baseline is placed at half screen height.
4170 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
4174 Show text for 1 second every 3 seconds:
4176 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:enable=lt(mod(t\,3)\,1):text='blink'"
4180 Use fontconfig to set the font. Note that the colons need to be escaped.
4182 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
4186 Print the date of a real-time encoding (see strftime(3)):
4188 drawtext='fontfile=FreeSans.ttf:text=%@{localtime:%a %b %d %Y@}'
4192 Show text fading in and out (appearing/disappearing):
4195 DS=1.0 # display start
4196 DE=10.0 # display end
4197 FID=1.5 # fade in duration
4198 FOD=5 # fade out duration
4199 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 @}"
4204 For more information about libfreetype, check:
4205 @url{http://www.freetype.org/}.
4207 For more information about fontconfig, check:
4208 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
4210 For more information about libfribidi, check:
4211 @url{http://fribidi.org/}.
4215 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
4217 The filter accepts the following options:
4222 Set low and high threshold values used by the Canny thresholding
4225 The high threshold selects the "strong" edge pixels, which are then
4226 connected through 8-connectivity with the "weak" edge pixels selected
4227 by the low threshold.
4229 @var{low} and @var{high} threshold values must be chosen in the range
4230 [0,1], and @var{low} should be lesser or equal to @var{high}.
4232 Default value for @var{low} is @code{20/255}, and default value for @var{high}
4236 Define the drawing mode.
4240 Draw white/gray wires on black background.
4243 Mix the colors to create a paint/cartoon effect.
4246 Default value is @var{wires}.
4249 @subsection Examples
4253 Standard edge detection with custom values for the hysteresis thresholding:
4255 edgedetect=low=0.1:high=0.4
4259 Painting effect without thresholding:
4261 edgedetect=mode=colormix:high=0
4265 @section extractplanes
4267 Extract color channel components from input video stream into
4268 separate grayscale video streams.
4270 The filter accepts the following option:
4274 Set plane(s) to extract.
4276 Available values for planes are:
4287 Choosing planes not available in the input will result in an error.
4288 That means you cannot select @code{r}, @code{g}, @code{b} planes
4289 with @code{y}, @code{u}, @code{v} planes at same time.
4292 @subsection Examples
4296 Extract luma, u and v color channel component from input video frame
4297 into 3 grayscale outputs:
4299 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
4305 Apply a posterize effect using the ELBG (Enhanced LBG) algorithm.
4307 For each input image, the filter will compute the optimal mapping from
4308 the input to the output given the codebook length, that is the number
4309 of distinct output colors.
4311 This filter accepts the following options.
4314 @item codebook_length, l
4315 Set codebook length. The value must be a positive integer, and
4316 represents the number of distinct output colors. Default value is 256.
4319 Set the maximum number of iterations to apply for computing the optimal
4320 mapping. The higher the value the better the result and the higher the
4321 computation time. Default value is 1.
4324 Set a random seed, must be an integer included between 0 and
4325 UINT32_MAX. If not specified, or if explicitly set to -1, the filter
4326 will try to use a good random seed on a best effort basis.
4331 Apply a fade-in/out effect to the input video.
4333 It accepts the following parameters:
4337 The effect type can be either "in" for a fade-in, or "out" for a fade-out
4339 Default is @code{in}.
4341 @item start_frame, s
4342 Specify the number of the frame to start applying the fade
4343 effect at. Default is 0.
4346 The number of frames that the fade effect lasts. At the end of the
4347 fade-in effect, the output video will have the same intensity as the input video.
4348 At the end of the fade-out transition, the output video will be filled with the
4349 selected @option{color}.
4353 If set to 1, fade only alpha channel, if one exists on the input.
4356 @item start_time, st
4357 Specify the timestamp (in seconds) of the frame to start to apply the fade
4358 effect. If both start_frame and start_time are specified, the fade will start at
4359 whichever comes last. Default is 0.
4362 The number of seconds for which the fade effect has to last. At the end of the
4363 fade-in effect the output video will have the same intensity as the input video,
4364 at the end of the fade-out transition the output video will be filled with the
4365 selected @option{color}.
4366 If both duration and nb_frames are specified, duration is used. Default is 0.
4369 Specify the color of the fade. Default is "black".
4372 @subsection Examples
4376 Fade in the first 30 frames of video:
4381 The command above is equivalent to:
4387 Fade out the last 45 frames of a 200-frame video:
4390 fade=type=out:start_frame=155:nb_frames=45
4394 Fade in the first 25 frames and fade out the last 25 frames of a 1000-frame video:
4396 fade=in:0:25, fade=out:975:25
4400 Make the first 5 frames yellow, then fade in from frame 5-24:
4402 fade=in:5:20:color=yellow
4406 Fade in alpha over first 25 frames of video:
4408 fade=in:0:25:alpha=1
4412 Make the first 5.5 seconds black, then fade in for 0.5 seconds:
4414 fade=t=in:st=5.5:d=0.5
4421 Extract a single field from an interlaced image using stride
4422 arithmetic to avoid wasting CPU time. The output frames are marked as
4425 The filter accepts the following options:
4429 Specify whether to extract the top (if the value is @code{0} or
4430 @code{top}) or the bottom field (if the value is @code{1} or
4436 Field matching filter for inverse telecine. It is meant to reconstruct the
4437 progressive frames from a telecined stream. The filter does not drop duplicated
4438 frames, so to achieve a complete inverse telecine @code{fieldmatch} needs to be
4439 followed by a decimation filter such as @ref{decimate} in the filtergraph.
4441 The separation of the field matching and the decimation is notably motivated by
4442 the possibility of inserting a de-interlacing filter fallback between the two.
4443 If the source has mixed telecined and real interlaced content,
4444 @code{fieldmatch} will not be able to match fields for the interlaced parts.
4445 But these remaining combed frames will be marked as interlaced, and thus can be
4446 de-interlaced by a later filter such as @ref{yadif} before decimation.
4448 In addition to the various configuration options, @code{fieldmatch} can take an
4449 optional second stream, activated through the @option{ppsrc} option. If
4450 enabled, the frames reconstruction will be based on the fields and frames from
4451 this second stream. This allows the first input to be pre-processed in order to
4452 help the various algorithms of the filter, while keeping the output lossless
4453 (assuming the fields are matched properly). Typically, a field-aware denoiser,
4454 or brightness/contrast adjustments can help.
4456 Note that this filter uses the same algorithms as TIVTC/TFM (AviSynth project)
4457 and VIVTC/VFM (VapourSynth project). The later is a light clone of TFM from
4458 which @code{fieldmatch} is based on. While the semantic and usage are very
4459 close, some behaviour and options names can differ.
4461 The filter accepts the following options:
4465 Specify the assumed field order of the input stream. Available values are:
4469 Auto detect parity (use FFmpeg's internal parity value).
4471 Assume bottom field first.
4473 Assume top field first.
4476 Note that it is sometimes recommended not to trust the parity announced by the
4479 Default value is @var{auto}.
4482 Set the matching mode or strategy to use. @option{pc} mode is the safest in the
4483 sense that it won't risk creating jerkiness due to duplicate frames when
4484 possible, but if there are bad edits or blended fields it will end up
4485 outputting combed frames when a good match might actually exist. On the other
4486 hand, @option{pcn_ub} mode is the most risky in terms of creating jerkiness,
4487 but will almost always find a good frame if there is one. The other values are
4488 all somewhere in between @option{pc} and @option{pcn_ub} in terms of risking
4489 jerkiness and creating duplicate frames versus finding good matches in sections
4490 with bad edits, orphaned fields, blended fields, etc.
4492 More details about p/c/n/u/b are available in @ref{p/c/n/u/b meaning} section.
4494 Available values are:
4498 2-way matching (p/c)
4500 2-way matching, and trying 3rd match if still combed (p/c + n)
4502 2-way matching, and trying 3rd match (same order) if still combed (p/c + u)
4504 2-way matching, trying 3rd match if still combed, and trying 4th/5th matches if
4505 still combed (p/c + n + u/b)
4507 3-way matching (p/c/n)
4509 3-way matching, and trying 4th/5th matches if all 3 of the original matches are
4510 detected as combed (p/c/n + u/b)
4513 The parenthesis at the end indicate the matches that would be used for that
4514 mode assuming @option{order}=@var{tff} (and @option{field} on @var{auto} or
4517 In terms of speed @option{pc} mode is by far the fastest and @option{pcn_ub} is
4520 Default value is @var{pc_n}.
4523 Mark the main input stream as a pre-processed input, and enable the secondary
4524 input stream as the clean source to pick the fields from. See the filter
4525 introduction for more details. It is similar to the @option{clip2} feature from
4528 Default value is @code{0} (disabled).
4531 Set the field to match from. It is recommended to set this to the same value as
4532 @option{order} unless you experience matching failures with that setting. In
4533 certain circumstances changing the field that is used to match from can have a
4534 large impact on matching performance. Available values are:
4538 Automatic (same value as @option{order}).
4540 Match from the bottom field.
4542 Match from the top field.
4545 Default value is @var{auto}.
4548 Set whether or not chroma is included during the match comparisons. In most
4549 cases it is recommended to leave this enabled. You should set this to @code{0}
4550 only if your clip has bad chroma problems such as heavy rainbowing or other
4551 artifacts. Setting this to @code{0} could also be used to speed things up at
4552 the cost of some accuracy.
4554 Default value is @code{1}.
4558 These define an exclusion band which excludes the lines between @option{y0} and
4559 @option{y1} from being included in the field matching decision. An exclusion
4560 band can be used to ignore subtitles, a logo, or other things that may
4561 interfere with the matching. @option{y0} sets the starting scan line and
4562 @option{y1} sets the ending line; all lines in between @option{y0} and
4563 @option{y1} (including @option{y0} and @option{y1}) will be ignored. Setting
4564 @option{y0} and @option{y1} to the same value will disable the feature.
4565 @option{y0} and @option{y1} defaults to @code{0}.
4568 Set the scene change detection threshold as a percentage of maximum change on
4569 the luma plane. Good values are in the @code{[8.0, 14.0]} range. Scene change
4570 detection is only relevant in case @option{combmatch}=@var{sc}. The range for
4571 @option{scthresh} is @code{[0.0, 100.0]}.
4573 Default value is @code{12.0}.
4576 When @option{combatch} is not @var{none}, @code{fieldmatch} will take into
4577 account the combed scores of matches when deciding what match to use as the
4578 final match. Available values are:
4582 No final matching based on combed scores.
4584 Combed scores are only used when a scene change is detected.
4586 Use combed scores all the time.
4589 Default is @var{sc}.
4592 Force @code{fieldmatch} to calculate the combed metrics for certain matches and
4593 print them. This setting is known as @option{micout} in TFM/VFM vocabulary.
4594 Available values are:
4598 No forced calculation.
4600 Force p/c/n calculations.
4602 Force p/c/n/u/b calculations.
4605 Default value is @var{none}.
4608 This is the area combing threshold used for combed frame detection. This
4609 essentially controls how "strong" or "visible" combing must be to be detected.
4610 Larger values mean combing must be more visible and smaller values mean combing
4611 can be less visible or strong and still be detected. Valid settings are from
4612 @code{-1} (every pixel will be detected as combed) to @code{255} (no pixel will
4613 be detected as combed). This is basically a pixel difference value. A good
4614 range is @code{[8, 12]}.
4616 Default value is @code{9}.
4619 Sets whether or not chroma is considered in the combed frame decision. Only
4620 disable this if your source has chroma problems (rainbowing, etc.) that are
4621 causing problems for the combed frame detection with chroma enabled. Actually,
4622 using @option{chroma}=@var{0} is usually more reliable, except for the case
4623 where there is chroma only combing in the source.
4625 Default value is @code{0}.
4629 Respectively set the x-axis and y-axis size of the window used during combed
4630 frame detection. This has to do with the size of the area in which
4631 @option{combpel} pixels are required to be detected as combed for a frame to be
4632 declared combed. See the @option{combpel} parameter description for more info.
4633 Possible values are any number that is a power of 2 starting at 4 and going up
4636 Default value is @code{16}.
4639 The number of combed pixels inside any of the @option{blocky} by
4640 @option{blockx} size blocks on the frame for the frame to be detected as
4641 combed. While @option{cthresh} controls how "visible" the combing must be, this
4642 setting controls "how much" combing there must be in any localized area (a
4643 window defined by the @option{blockx} and @option{blocky} settings) on the
4644 frame. Minimum value is @code{0} and maximum is @code{blocky x blockx} (at
4645 which point no frames will ever be detected as combed). This setting is known
4646 as @option{MI} in TFM/VFM vocabulary.
4648 Default value is @code{80}.
4651 @anchor{p/c/n/u/b meaning}
4652 @subsection p/c/n/u/b meaning
4654 @subsubsection p/c/n
4656 We assume the following telecined stream:
4659 Top fields: 1 2 2 3 4
4660 Bottom fields: 1 2 3 4 4
4663 The numbers correspond to the progressive frame the fields relate to. Here, the
4664 first two frames are progressive, the 3rd and 4th are combed, and so on.
4666 When @code{fieldmatch} is configured to run a matching from bottom
4667 (@option{field}=@var{bottom}) this is how this input stream get transformed:
4672 B 1 2 3 4 4 <-- matching reference
4681 As a result of the field matching, we can see that some frames get duplicated.
4682 To perform a complete inverse telecine, you need to rely on a decimation filter
4683 after this operation. See for instance the @ref{decimate} filter.
4685 The same operation now matching from top fields (@option{field}=@var{top})
4690 T 1 2 2 3 4 <-- matching reference
4700 In these examples, we can see what @var{p}, @var{c} and @var{n} mean;
4701 basically, they refer to the frame and field of the opposite parity:
4704 @item @var{p} matches the field of the opposite parity in the previous frame
4705 @item @var{c} matches the field of the opposite parity in the current frame
4706 @item @var{n} matches the field of the opposite parity in the next frame
4711 The @var{u} and @var{b} matching are a bit special in the sense that they match
4712 from the opposite parity flag. In the following examples, we assume that we are
4713 currently matching the 2nd frame (Top:2, bottom:2). According to the match, a
4714 'x' is placed above and below each matched fields.
4716 With bottom matching (@option{field}=@var{bottom}):
4721 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
4722 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
4730 With top matching (@option{field}=@var{top}):
4735 Top 1 2 2 1 2 2 1 2 2 1 2 2 1 2 2
4736 Bottom 1 2 3 1 2 3 1 2 3 1 2 3 1 2 3
4744 @subsection Examples
4746 Simple IVTC of a top field first telecined stream:
4748 fieldmatch=order=tff:combmatch=none, decimate
4751 Advanced IVTC, with fallback on @ref{yadif} for still combed frames:
4753 fieldmatch=order=tff:combmatch=full, yadif=deint=interlaced, decimate
4758 Transform the field order of the input video.
4760 It accepts the following parameters:
4765 The output field order. Valid values are @var{tff} for top field first or @var{bff}
4766 for bottom field first.
4769 The default value is @samp{tff}.
4771 The transformation is done by shifting the picture content up or down
4772 by one line, and filling the remaining line with appropriate picture content.
4773 This method is consistent with most broadcast field order converters.
4775 If the input video is not flagged as being interlaced, or it is already
4776 flagged as being of the required output field order, then this filter does
4777 not alter the incoming video.
4779 It is very useful when converting to or from PAL DV material,
4780 which is bottom field first.
4784 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
4789 Buffer input images and send them when they are requested.
4791 It is mainly useful when auto-inserted by the libavfilter
4794 It does not take parameters.
4799 Convert the input video to one of the specified pixel formats.
4800 Libavfilter will try to pick one that is suitable as input to
4803 It accepts the following parameters:
4807 A '|'-separated list of pixel format names, such as
4808 "pix_fmts=yuv420p|monow|rgb24".
4812 @subsection Examples
4816 Convert the input video to the @var{yuv420p} format
4818 format=pix_fmts=yuv420p
4821 Convert the input video to any of the formats in the list
4823 format=pix_fmts=yuv420p|yuv444p|yuv410p
4830 Convert the video to specified constant frame rate by duplicating or dropping
4831 frames as necessary.
4833 It accepts the following parameters:
4837 The desired output frame rate. The default is @code{25}.
4842 Possible values are:
4845 zero round towards 0
4849 round towards -infinity
4851 round towards +infinity
4855 The default is @code{near}.
4858 Assume the first PTS should be the given value, in seconds. This allows for
4859 padding/trimming at the start of stream. By default, no assumption is made
4860 about the first frame's expected PTS, so no padding or trimming is done.
4861 For example, this could be set to 0 to pad the beginning with duplicates of
4862 the first frame if a video stream starts after the audio stream or to trim any
4863 frames with a negative PTS.
4867 Alternatively, the options can be specified as a flat string:
4868 @var{fps}[:@var{round}].
4870 See also the @ref{setpts} filter.
4872 @subsection Examples
4876 A typical usage in order to set the fps to 25:
4882 Sets the fps to 24, using abbreviation and rounding method to round to nearest:
4884 fps=fps=film:round=near
4890 Pack two different video streams into a stereoscopic video, setting proper
4891 metadata on supported codecs. The two views should have the same size and
4892 framerate and processing will stop when the shorter video ends. Please note
4893 that you may conveniently adjust view properties with the @ref{scale} and
4896 It accepts the following parameters:
4900 The desired packing format. Supported values are:
4905 The views are next to each other (default).
4908 The views are on top of each other.
4911 The views are packed by line.
4914 The views are packed by column.
4917 The views are temporally interleaved.
4926 # Convert left and right views into a frame-sequential video
4927 ffmpeg -i LEFT -i RIGHT -filter_complex framepack=frameseq OUTPUT
4929 # Convert views into a side-by-side video with the same output resolution as the input
4930 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
4935 Select one frame every N-th frame.
4937 This filter accepts the following option:
4940 Select frame after every @code{step} frames.
4941 Allowed values are positive integers higher than 0. Default value is @code{1}.
4947 Apply a frei0r effect to the input video.
4949 To enable the compilation of this filter, you need to install the frei0r
4950 header and configure FFmpeg with @code{--enable-frei0r}.
4952 It accepts the following parameters:
4957 The name of the frei0r effect to load. If the environment variable
4958 @env{FREI0R_PATH} is defined, the frei0r effect is searched for in each of the
4959 directories specified by the colon-separated list in @env{FREIOR_PATH}.
4960 Otherwise, the standard frei0r paths are searched, in this order:
4961 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
4962 @file{/usr/lib/frei0r-1/}.
4965 A '|'-separated list of parameters to pass to the frei0r effect.
4969 A frei0r effect parameter can be a boolean (its value is either
4970 "y" or "n"), a double, a color (specified as
4971 @var{R}/@var{G}/@var{B}, where @var{R}, @var{G}, and @var{B} are floating point
4972 numbers between 0.0 and 1.0, inclusive) or by a color description specified in the "Color"
4973 section in the ffmpeg-utils manual), a position (specified as @var{X}/@var{Y}, where
4974 @var{X} and @var{Y} are floating point numbers) and/or a string.
4976 The number and types of parameters depend on the loaded effect. If an
4977 effect parameter is not specified, the default value is set.
4979 @subsection Examples
4983 Apply the distort0r effect, setting the first two double parameters:
4985 frei0r=filter_name=distort0r:filter_params=0.5|0.01
4989 Apply the colordistance effect, taking a color as the first parameter:
4991 frei0r=colordistance:0.2/0.3/0.4
4992 frei0r=colordistance:violet
4993 frei0r=colordistance:0x112233
4997 Apply the perspective effect, specifying the top left and top right image
5000 frei0r=perspective:0.2/0.2|0.8/0.2
5004 For more information, see
5005 @url{http://frei0r.dyne.org}
5009 The filter accepts the following options:
5013 Set the luminance expression.
5015 Set the chrominance blue expression.
5017 Set the chrominance red expression.
5019 Set the alpha expression.
5021 Set the red expression.
5023 Set the green expression.
5025 Set the blue expression.
5028 The colorspace is selected according to the specified options. If one
5029 of the @option{lum_expr}, @option{cb_expr}, or @option{cr_expr}
5030 options is specified, the filter will automatically select a YCbCr
5031 colorspace. If one of the @option{red_expr}, @option{green_expr}, or
5032 @option{blue_expr} options is specified, it will select an RGB
5035 If one of the chrominance expression is not defined, it falls back on the other
5036 one. If no alpha expression is specified it will evaluate to opaque value.
5037 If none of chrominance expressions are specified, they will evaluate
5038 to the luminance expression.
5040 The expressions can use the following variables and functions:
5044 The sequential number of the filtered frame, starting from @code{0}.
5048 The coordinates of the current sample.
5052 The width and height of the image.
5056 Width and height scale depending on the currently filtered plane. It is the
5057 ratio between the corresponding luma plane number of pixels and the current
5058 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
5059 @code{0.5,0.5} for chroma planes.
5062 Time of the current frame, expressed in seconds.
5065 Return the value of the pixel at location (@var{x},@var{y}) of the current
5069 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
5073 Return the value of the pixel at location (@var{x},@var{y}) of the
5074 blue-difference chroma plane. Return 0 if there is no such plane.
5077 Return the value of the pixel at location (@var{x},@var{y}) of the
5078 red-difference chroma plane. Return 0 if there is no such plane.
5083 Return the value of the pixel at location (@var{x},@var{y}) of the
5084 red/green/blue component. Return 0 if there is no such component.
5087 Return the value of the pixel at location (@var{x},@var{y}) of the alpha
5088 plane. Return 0 if there is no such plane.
5091 For functions, if @var{x} and @var{y} are outside the area, the value will be
5092 automatically clipped to the closer edge.
5094 @subsection Examples
5098 Flip the image horizontally:
5104 Generate a bidimensional sine wave, with angle @code{PI/3} and a
5105 wavelength of 100 pixels:
5107 geq=128 + 100*sin(2*(PI/100)*(cos(PI/3)*(X-50*T) + sin(PI/3)*Y)):128:128
5111 Generate a fancy enigmatic moving light:
5113 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
5117 Generate a quick emboss effect:
5119 format=gray,geq=lum_expr='(p(X,Y)+(256-p(X-4,Y-4)))/2'
5123 Modify RGB components depending on pixel position:
5125 geq=r='X/W*r(X,Y)':g='(1-X/W)*g(X,Y)':b='(H-Y)/H*b(X,Y)'
5131 Fix the banding artifacts that are sometimes introduced into nearly flat
5132 regions by truncation to 8bit color depth.
5133 Interpolate the gradients that should go where the bands are, and
5136 It is designed for playback only. Do not use it prior to
5137 lossy compression, because compression tends to lose the dither and
5138 bring back the bands.
5140 It accepts the following parameters:
5145 The maximum amount by which the filter will change any one pixel. This is also
5146 the threshold for detecting nearly flat regions. Acceptable values range from
5147 .51 to 64; the default value is 1.2. Out-of-range values will be clipped to the
5151 The neighborhood to fit the gradient to. A larger radius makes for smoother
5152 gradients, but also prevents the filter from modifying the pixels near detailed
5153 regions. Acceptable values are 8-32; the default value is 16. Out-of-range
5154 values will be clipped to the valid range.
5158 Alternatively, the options can be specified as a flat string:
5159 @var{strength}[:@var{radius}]
5161 @subsection Examples
5165 Apply the filter with a @code{3.5} strength and radius of @code{8}:
5171 Specify radius, omitting the strength (which will fall-back to the default
5182 Apply a Hald CLUT to a video stream.
5184 First input is the video stream to process, and second one is the Hald CLUT.
5185 The Hald CLUT input can be a simple picture or a complete video stream.
5187 The filter accepts the following options:
5191 Force termination when the shortest input terminates. Default is @code{0}.
5193 Continue applying the last CLUT after the end of the stream. A value of
5194 @code{0} disable the filter after the last frame of the CLUT is reached.
5195 Default is @code{1}.
5198 @code{haldclut} also has the same interpolation options as @ref{lut3d} (both
5199 filters share the same internals).
5201 More information about the Hald CLUT can be found on Eskil Steenberg's website
5202 (Hald CLUT author) at @url{http://www.quelsolaar.com/technology/clut.html}.
5204 @subsection Workflow examples
5206 @subsubsection Hald CLUT video stream
5208 Generate an identity Hald CLUT stream altered with various effects:
5210 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
5213 Note: make sure you use a lossless codec.
5215 Then use it with @code{haldclut} to apply it on some random stream:
5217 ffmpeg -f lavfi -i mandelbrot -i clut.nut -filter_complex '[0][1] haldclut' -t 20 mandelclut.mkv
5220 The Hald CLUT will be applied to the 10 first seconds (duration of
5221 @file{clut.nut}), then the latest picture of that CLUT stream will be applied
5222 to the remaining frames of the @code{mandelbrot} stream.
5224 @subsubsection Hald CLUT with preview
5226 A Hald CLUT is supposed to be a squared image of @code{Level*Level*Level} by
5227 @code{Level*Level*Level} pixels. For a given Hald CLUT, FFmpeg will select the
5228 biggest possible square starting at the top left of the picture. The remaining
5229 padding pixels (bottom or right) will be ignored. This area can be used to add
5230 a preview of the Hald CLUT.
5232 Typically, the following generated Hald CLUT will be supported by the
5233 @code{haldclut} filter:
5236 ffmpeg -f lavfi -i @ref{haldclutsrc}=8 -vf "
5237 pad=iw+320 [padded_clut];
5238 smptebars=s=320x256, split [a][b];
5239 [padded_clut][a] overlay=W-320:h, curves=color_negative [main];
5240 [main][b] overlay=W-320" -frames:v 1 clut.png
5243 It contains the original and a preview of the effect of the CLUT: SMPTE color
5244 bars are displayed on the right-top, and below the same color bars processed by
5247 Then, the effect of this Hald CLUT can be visualized with:
5249 ffplay input.mkv -vf "movie=clut.png, [in] haldclut"
5254 Flip the input video horizontally.
5256 For example, to horizontally flip the input video with @command{ffmpeg}:
5258 ffmpeg -i in.avi -vf "hflip" out.avi
5262 This filter applies a global color histogram equalization on a
5265 It can be used to correct video that has a compressed range of pixel
5266 intensities. The filter redistributes the pixel intensities to
5267 equalize their distribution across the intensity range. It may be
5268 viewed as an "automatically adjusting contrast filter". This filter is
5269 useful only for correcting degraded or poorly captured source
5272 The filter accepts the following options:
5276 Determine the amount of equalization to be applied. As the strength
5277 is reduced, the distribution of pixel intensities more-and-more
5278 approaches that of the input frame. The value must be a float number
5279 in the range [0,1] and defaults to 0.200.
5282 Set the maximum intensity that can generated and scale the output
5283 values appropriately. The strength should be set as desired and then
5284 the intensity can be limited if needed to avoid washing-out. The value
5285 must be a float number in the range [0,1] and defaults to 0.210.
5288 Set the antibanding level. If enabled the filter will randomly vary
5289 the luminance of output pixels by a small amount to avoid banding of
5290 the histogram. Possible values are @code{none}, @code{weak} or
5291 @code{strong}. It defaults to @code{none}.
5296 Compute and draw a color distribution histogram for the input video.
5298 The computed histogram is a representation of the color component
5299 distribution in an image.
5301 The filter accepts the following options:
5307 It accepts the following values:
5310 Standard histogram that displays the color components distribution in an
5311 image. Displays color graph for each color component. Shows distribution of
5312 the Y, U, V, A or R, G, B components, depending on input format, in the
5313 current frame. Below each graph a color component scale meter is shown.
5316 Displays chroma values (U/V color placement) in a two dimensional
5317 graph (which is called a vectorscope). The brighter a pixel in the
5318 vectorscope, the more pixels of the input frame correspond to that pixel
5319 (i.e., more pixels have this chroma value). The V component is displayed on
5320 the horizontal (X) axis, with the leftmost side being V = 0 and the rightmost
5321 side being V = 255. The U component is displayed on the vertical (Y) axis,
5322 with the top representing U = 0 and the bottom representing U = 255.
5324 The position of a white pixel in the graph corresponds to the chroma value of
5325 a pixel of the input clip. The graph can therefore be used to read the hue
5326 (color flavor) and the saturation (the dominance of the hue in the color). As
5327 the hue of a color changes, it moves around the square. At the center of the
5328 square the saturation is zero, which means that the corresponding pixel has no
5329 color. If the amount of a specific color is increased (while leaving the other
5330 colors unchanged) the saturation increases, and the indicator moves towards
5331 the edge of the square.
5334 Chroma values in vectorscope, similar as @code{color} but actual chroma values
5338 Per row/column color component graph. In row mode, the graph on the left side
5339 represents color component value 0 and the right side represents value = 255.
5340 In column mode, the top side represents color component value = 0 and bottom
5341 side represents value = 255.
5343 Default value is @code{levels}.
5346 Set height of level in @code{levels}. Default value is @code{200}.
5347 Allowed range is [50, 2048].
5350 Set height of color scale in @code{levels}. Default value is @code{12}.
5351 Allowed range is [0, 40].
5354 Set step for @code{waveform} mode. Smaller values are useful to find out how
5355 many values of the same luminance are distributed across input rows/columns.
5356 Default value is @code{10}. Allowed range is [1, 255].
5359 Set mode for @code{waveform}. Can be either @code{row}, or @code{column}.
5360 Default is @code{row}.
5362 @item waveform_mirror
5363 Set mirroring mode for @code{waveform}. @code{0} means unmirrored, @code{1}
5364 means mirrored. In mirrored mode, higher values will be represented on the left
5365 side for @code{row} mode and at the top for @code{column} mode. Default is
5366 @code{0} (unmirrored).
5369 Set display mode for @code{waveform} and @code{levels}.
5370 It accepts the following values:
5373 Display separate graph for the color components side by side in
5374 @code{row} waveform mode or one below the other in @code{column} waveform mode
5375 for @code{waveform} histogram mode. For @code{levels} histogram mode,
5376 per color component graphs are placed below each other.
5378 Using this display mode in @code{waveform} histogram mode makes it easy to
5379 spot color casts in the highlights and shadows of an image, by comparing the
5380 contours of the top and the bottom graphs of each waveform. Since whites,
5381 grays, and blacks are characterized by exactly equal amounts of red, green,
5382 and blue, neutral areas of the picture should display three waveforms of
5383 roughly equal width/height. If not, the correction is easy to perform by
5384 making level adjustments the three waveforms.
5387 Presents information identical to that in the @code{parade}, except
5388 that the graphs representing color components are superimposed directly
5391 This display mode in @code{waveform} histogram mode makes it easier to spot
5392 relative differences or similarities in overlapping areas of the color
5393 components that are supposed to be identical, such as neutral whites, grays,
5396 Default is @code{parade}.
5399 Set mode for @code{levels}. Can be either @code{linear}, or @code{logarithmic}.
5400 Default is @code{linear}.
5403 @subsection Examples
5408 Calculate and draw histogram:
5410 ffplay -i input -vf histogram
5418 This is a high precision/quality 3d denoise filter. It aims to reduce
5419 image noise, producing smooth images and making still images really
5420 still. It should enhance compressibility.
5422 It accepts the following optional parameters:
5426 A non-negative floating point number which specifies spatial luma strength.
5429 @item chroma_spatial
5430 A non-negative floating point number which specifies spatial chroma strength.
5431 It defaults to 3.0*@var{luma_spatial}/4.0.
5434 A floating point number which specifies luma temporal strength. It defaults to
5435 6.0*@var{luma_spatial}/4.0.
5438 A floating point number which specifies chroma temporal strength. It defaults to
5439 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}.
5444 Apply a high-quality magnification filter designed for pixel art. This filter
5445 was originally created by Maxim Stepin.
5447 It accepts the following option:
5451 Set the scaling dimension: @code{2} for @code{hq2x}, @code{3} for
5452 @code{hq3x} and @code{4} for @code{hq4x}.
5453 Default is @code{3}.
5458 Modify the hue and/or the saturation of the input.
5460 It accepts the following parameters:
5464 Specify the hue angle as a number of degrees. It accepts an expression,
5465 and defaults to "0".
5468 Specify the saturation in the [-10,10] range. It accepts an expression and
5472 Specify the hue angle as a number of radians. It accepts an
5473 expression, and defaults to "0".
5476 Specify the brightness in the [-10,10] range. It accepts an expression and
5480 @option{h} and @option{H} are mutually exclusive, and can't be
5481 specified at the same time.
5483 The @option{b}, @option{h}, @option{H} and @option{s} option values are
5484 expressions containing the following constants:
5488 frame count of the input frame starting from 0
5491 presentation timestamp of the input frame expressed in time base units
5494 frame rate of the input video, NAN if the input frame rate is unknown
5497 timestamp expressed in seconds, NAN if the input timestamp is unknown
5500 time base of the input video
5503 @subsection Examples
5507 Set the hue to 90 degrees and the saturation to 1.0:
5513 Same command but expressing the hue in radians:
5519 Rotate hue and make the saturation swing between 0
5520 and 2 over a period of 1 second:
5522 hue="H=2*PI*t: s=sin(2*PI*t)+1"
5526 Apply a 3 seconds saturation fade-in effect starting at 0:
5531 The general fade-in expression can be written as:
5533 hue="s=min(0\, max((t-START)/DURATION\, 1))"
5537 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
5539 hue="s=max(0\, min(1\, (8-t)/3))"
5542 The general fade-out expression can be written as:
5544 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
5549 @subsection Commands
5551 This filter supports the following commands:
5557 Modify the hue and/or the saturation and/or brightness of the input video.
5558 The command accepts the same syntax of the corresponding option.
5560 If the specified expression is not valid, it is kept at its current
5566 Detect video interlacing type.
5568 This filter tries to detect if the input is interlaced or progressive,
5569 top or bottom field first.
5571 The filter accepts the following options:
5575 Set interlacing threshold.
5577 Set progressive threshold.
5582 Deinterleave or interleave fields.
5584 This filter allows one to process interlaced images fields without
5585 deinterlacing them. Deinterleaving splits the input frame into 2
5586 fields (so called half pictures). Odd lines are moved to the top
5587 half of the output image, even lines to the bottom half.
5588 You can process (filter) them independently and then re-interleave them.
5590 The filter accepts the following options:
5594 @item chroma_mode, c
5596 Available values for @var{luma_mode}, @var{chroma_mode} and
5597 @var{alpha_mode} are:
5603 @item deinterleave, d
5604 Deinterleave fields, placing one above the other.
5607 Interleave fields. Reverse the effect of deinterleaving.
5609 Default value is @code{none}.
5612 @item chroma_swap, cs
5613 @item alpha_swap, as
5614 Swap luma/chroma/alpha fields. Exchange even & odd lines. Default value is @code{0}.
5619 Simple interlacing filter from progressive contents. This interleaves upper (or
5620 lower) lines from odd frames with lower (or upper) lines from even frames,
5621 halving the frame rate and preserving image height.
5624 Original Original New Frame
5625 Frame 'j' Frame 'j+1' (tff)
5626 ========== =========== ==================
5627 Line 0 --------------------> Frame 'j' Line 0
5628 Line 1 Line 1 ----> Frame 'j+1' Line 1
5629 Line 2 ---------------------> Frame 'j' Line 2
5630 Line 3 Line 3 ----> Frame 'j+1' Line 3
5632 New Frame + 1 will be generated by Frame 'j+2' and Frame 'j+3' and so on
5635 It accepts the following optional parameters:
5639 This determines whether the interlaced frame is taken from the even
5640 (tff - default) or odd (bff) lines of the progressive frame.
5643 Enable (default) or disable the vertical lowpass filter to avoid twitter
5644 interlacing and reduce moire patterns.
5649 Deinterlace input video by applying Donald Graft's adaptive kernel
5650 deinterling. Work on interlaced parts of a video to produce
5653 The description of the accepted parameters follows.
5657 Set the threshold which affects the filter's tolerance when
5658 determining if a pixel line must be processed. It must be an integer
5659 in the range [0,255] and defaults to 10. A value of 0 will result in
5660 applying the process on every pixels.
5663 Paint pixels exceeding the threshold value to white if set to 1.
5667 Set the fields order. Swap fields if set to 1, leave fields alone if
5671 Enable additional sharpening if set to 1. Default is 0.
5674 Enable twoway sharpening if set to 1. Default is 0.
5677 @subsection Examples
5681 Apply default values:
5683 kerndeint=thresh=10:map=0:order=0:sharp=0:twoway=0
5687 Enable additional sharpening:
5693 Paint processed pixels in white:
5699 @section lenscorrection
5701 Correct radial lens distortion
5703 This filter can be used to correct for radial distortion as can result from the use
5704 of wide angle lenses, and thereby re-rectify the image. To find the right parameters
5705 one can use tools available for example as part of opencv or simply trial-and-error.
5706 To use opencv use the calibration sample (under samples/cpp) from the opencv sources
5707 and extract the k1 and k2 coefficients from the resulting matrix.
5709 Note that effectively the same filter is available in the open-source tools Krita and
5710 Digikam from the KDE project.
5712 In contrast to the @ref{vignette} filter, which can also be used to compensate lens errors,
5713 this filter corrects the distortion of the image, whereas @ref{vignette} corrects the
5714 brightness distribution, so you may want to use both filters together in certain
5715 cases, though you will have to take care of ordering, i.e. whether vignetting should
5716 be applied before or after lens correction.
5720 The filter accepts the following options:
5724 Relative x-coordinate of the focal point of the image, and thereby the center of the
5725 distrortion. This value has a range [0,1] and is expressed as fractions of the image
5728 Relative y-coordinate of the focal point of the image, and thereby the center of the
5729 distrortion. This value has a range [0,1] and is expressed as fractions of the image
5732 Coefficient of the quadratic correction term. 0.5 means no correction.
5734 Coefficient of the double quadratic correction term. 0.5 means no correction.
5737 The formula that generates the correction is:
5739 @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)
5741 where @var{r_0} is halve of the image diagonal and @var{r_src} and @var{r_tgt} are the
5742 distances from the focal point in the source and target images, respectively.
5747 Apply a 3D LUT to an input video.
5749 The filter accepts the following options:
5753 Set the 3D LUT file name.
5755 Currently supported formats:
5767 Select interpolation mode.
5769 Available values are:
5773 Use values from the nearest defined point.
5775 Interpolate values using the 8 points defining a cube.
5777 Interpolate values using a tetrahedron.
5781 @section lut, lutrgb, lutyuv
5783 Compute a look-up table for binding each pixel component input value
5784 to an output value, and apply it to the input video.
5786 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
5787 to an RGB input video.
5789 These filters accept the following parameters:
5792 set first pixel component expression
5794 set second pixel component expression
5796 set third pixel component expression
5798 set fourth pixel component expression, corresponds to the alpha component
5801 set red component expression
5803 set green component expression
5805 set blue component expression
5807 alpha component expression
5810 set Y/luminance component expression
5812 set U/Cb component expression
5814 set V/Cr component expression
5817 Each of them specifies the expression to use for computing the lookup table for
5818 the corresponding pixel component values.
5820 The exact component associated to each of the @var{c*} options depends on the
5823 The @var{lut} filter requires either YUV or RGB pixel formats in input,
5824 @var{lutrgb} requires RGB pixel formats in input, and @var{lutyuv} requires YUV.
5826 The expressions can contain the following constants and functions:
5831 The input width and height.
5834 The input value for the pixel component.
5837 The input value, clipped to the @var{minval}-@var{maxval} range.
5840 The maximum value for the pixel component.
5843 The minimum value for the pixel component.
5846 The negated value for the pixel component value, clipped to the
5847 @var{minval}-@var{maxval} range; it corresponds to the expression
5848 "maxval-clipval+minval".
5851 The computed value in @var{val}, clipped to the
5852 @var{minval}-@var{maxval} range.
5854 @item gammaval(gamma)
5855 The computed gamma correction value of the pixel component value,
5856 clipped to the @var{minval}-@var{maxval} range. It corresponds to the
5858 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
5862 All expressions default to "val".
5864 @subsection Examples
5870 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
5871 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
5874 The above is the same as:
5876 lutrgb="r=negval:g=negval:b=negval"
5877 lutyuv="y=negval:u=negval:v=negval"
5887 Remove chroma components, turning the video into a graytone image:
5889 lutyuv="u=128:v=128"
5893 Apply a luma burning effect:
5899 Remove green and blue components:
5905 Set a constant alpha channel value on input:
5907 format=rgba,lutrgb=a="maxval-minval/2"
5911 Correct luminance gamma by a factor of 0.5:
5913 lutyuv=y=gammaval(0.5)
5917 Discard least significant bits of luma:
5919 lutyuv=y='bitand(val, 128+64+32)'
5923 @section mergeplanes
5925 Merge color channel components from several video streams.
5927 The filter accepts up to 4 input streams, and merge selected input
5928 planes to the output video.
5930 This filter accepts the following options:
5933 Set input to output plane mapping. Default is @code{0}.
5935 The mappings is specified as a bitmap. It should be specified as a
5936 hexadecimal number in the form 0xAa[Bb[Cc[Dd]]]. 'Aa' describes the
5937 mapping for the first plane of the output stream. 'A' sets the number of
5938 the input stream to use (from 0 to 3), and 'a' the plane number of the
5939 corresponding input to use (from 0 to 3). The rest of the mappings is
5940 similar, 'Bb' describes the mapping for the output stream second
5941 plane, 'Cc' describes the mapping for the output stream third plane and
5942 'Dd' describes the mapping for the output stream fourth plane.
5945 Set output pixel format. Default is @code{yuva444p}.
5948 @subsection Examples
5952 Merge three gray video streams of same width and height into single video stream:
5954 [a0][a1][a2]mergeplanes=0x001020:yuv444p
5958 Merge 1st yuv444p stream and 2nd gray video stream into yuva444p video stream:
5960 [a0][a1]mergeplanes=0x00010210:yuva444p
5964 Swap Y and A plane in yuva444p stream:
5966 format=yuva444p,mergeplanes=0x03010200:yuva444p
5970 Swap U and V plane in yuv420p stream:
5972 format=yuv420p,mergeplanes=0x000201:yuv420p
5976 Cast a rgb24 clip to yuv444p:
5978 format=rgb24,mergeplanes=0x000102:yuv444p
5984 Apply motion-compensation deinterlacing.
5986 It needs one field per frame as input and must thus be used together
5987 with yadif=1/3 or equivalent.
5989 This filter accepts the following options:
5992 Set the deinterlacing mode.
5994 It accepts one of the following values:
5999 use iterative motion estimation
6001 like @samp{slow}, but use multiple reference frames.
6003 Default value is @samp{fast}.
6006 Set the picture field parity assumed for the input video. It must be
6007 one of the following values:
6011 assume top field first
6013 assume bottom field first
6016 Default value is @samp{bff}.
6019 Set per-block quantization parameter (QP) used by the internal
6022 Higher values should result in a smoother motion vector field but less
6023 optimal individual vectors. Default value is 1.
6028 Apply an MPlayer filter to the input video.
6030 This filter provides a wrapper around some of the filters of
6033 This wrapper is considered experimental. Some of the wrapped filters
6034 may not work properly and we may drop support for them, as they will
6035 be implemented natively into FFmpeg. Thus you should avoid
6036 depending on them when writing portable scripts.
6038 The filter accepts the parameters:
6039 @var{filter_name}[:=]@var{filter_params}
6041 @var{filter_name} is the name of a supported MPlayer filter,
6042 @var{filter_params} is a string containing the parameters accepted by
6045 The list of the currently supported filters follows:
6056 The parameter syntax and behavior for the listed filters are the same
6057 of the corresponding MPlayer filters. For detailed instructions check
6058 the "VIDEO FILTERS" section in the MPlayer manual.
6060 @subsection Examples
6064 Adjust gamma, brightness, contrast:
6070 See also mplayer(1), @url{http://www.mplayerhq.hu/}.
6074 Drop frames that do not differ greatly from the previous frame in
6075 order to reduce frame rate.
6077 The main use of this filter is for very-low-bitrate encoding
6078 (e.g. streaming over dialup modem), but it could in theory be used for
6079 fixing movies that were inverse-telecined incorrectly.
6081 A description of the accepted options follows.
6085 Set the maximum number of consecutive frames which can be dropped (if
6086 positive), or the minimum interval between dropped frames (if
6087 negative). If the value is 0, the frame is dropped unregarding the
6088 number of previous sequentially dropped frames.
6095 Set the dropping threshold values.
6097 Values for @option{hi} and @option{lo} are for 8x8 pixel blocks and
6098 represent actual pixel value differences, so a threshold of 64
6099 corresponds to 1 unit of difference for each pixel, or the same spread
6100 out differently over the block.
6102 A frame is a candidate for dropping if no 8x8 blocks differ by more
6103 than a threshold of @option{hi}, and if no more than @option{frac} blocks (1
6104 meaning the whole image) differ by more than a threshold of @option{lo}.
6106 Default value for @option{hi} is 64*12, default value for @option{lo} is
6107 64*5, and default value for @option{frac} is 0.33.
6115 It accepts an integer in input; if non-zero it negates the
6116 alpha component (if available). The default value in input is 0.
6120 Force libavfilter not to use any of the specified pixel formats for the
6121 input to the next filter.
6123 It accepts the following parameters:
6127 A '|'-separated list of pixel format names, such as
6128 apix_fmts=yuv420p|monow|rgb24".
6132 @subsection Examples
6136 Force libavfilter to use a format different from @var{yuv420p} for the
6137 input to the vflip filter:
6139 noformat=pix_fmts=yuv420p,vflip
6143 Convert the input video to any of the formats not contained in the list:
6145 noformat=yuv420p|yuv444p|yuv410p
6151 Add noise on video input frame.
6153 The filter accepts the following options:
6161 Set noise seed for specific pixel component or all pixel components in case
6162 of @var{all_seed}. Default value is @code{123457}.
6164 @item all_strength, alls
6165 @item c0_strength, c0s
6166 @item c1_strength, c1s
6167 @item c2_strength, c2s
6168 @item c3_strength, c3s
6169 Set noise strength for specific pixel component or all pixel components in case
6170 @var{all_strength}. Default value is @code{0}. Allowed range is [0, 100].
6172 @item all_flags, allf
6177 Set pixel component flags or set flags for all components if @var{all_flags}.
6178 Available values for component flags are:
6181 averaged temporal noise (smoother)
6183 mix random noise with a (semi)regular pattern
6185 temporal noise (noise pattern changes between frames)
6187 uniform noise (gaussian otherwise)
6191 @subsection Examples
6193 Add temporal and uniform noise to input video:
6195 noise=alls=20:allf=t+u
6200 Pass the video source unchanged to the output.
6204 Apply a video transform using libopencv.
6206 To enable this filter, install the libopencv library and headers and
6207 configure FFmpeg with @code{--enable-libopencv}.
6209 It accepts the following parameters:
6214 The name of the libopencv filter to apply.
6217 The parameters to pass to the libopencv filter. If not specified, the default
6222 Refer to the official libopencv documentation for more precise
6224 @url{http://opencv.willowgarage.com/documentation/c/image_filtering.html}
6226 Several libopencv filters are supported; see the following subsections.
6231 Dilate an image by using a specific structuring element.
6232 It corresponds to the libopencv function @code{cvDilate}.
6234 It accepts the parameters: @var{struct_el}|@var{nb_iterations}.
6236 @var{struct_el} represents a structuring element, and has the syntax:
6237 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
6239 @var{cols} and @var{rows} represent the number of columns and rows of
6240 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
6241 point, and @var{shape} the shape for the structuring element. @var{shape}
6242 must be "rect", "cross", "ellipse", or "custom".
6244 If the value for @var{shape} is "custom", it must be followed by a
6245 string of the form "=@var{filename}". The file with name
6246 @var{filename} is assumed to represent a binary image, with each
6247 printable character corresponding to a bright pixel. When a custom
6248 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
6249 or columns and rows of the read file are assumed instead.
6251 The default value for @var{struct_el} is "3x3+0x0/rect".
6253 @var{nb_iterations} specifies the number of times the transform is
6254 applied to the image, and defaults to 1.
6258 # Use the default values
6261 # Dilate using a structuring element with a 5x5 cross, iterating two times
6262 ocv=filter_name=dilate:filter_params=5x5+2x2/cross|2
6264 # Read the shape from the file diamond.shape, iterating two times.
6265 # The file diamond.shape may contain a pattern of characters like this
6271 # The specified columns and rows are ignored
6272 # but the anchor point coordinates are not
6273 ocv=dilate:0x0+2x2/custom=diamond.shape|2
6278 Erode an image by using a specific structuring element.
6279 It corresponds to the libopencv function @code{cvErode}.
6281 It accepts the parameters: @var{struct_el}:@var{nb_iterations},
6282 with the same syntax and semantics as the @ref{dilate} filter.
6286 Smooth the input video.
6288 The filter takes the following parameters:
6289 @var{type}|@var{param1}|@var{param2}|@var{param3}|@var{param4}.
6291 @var{type} is the type of smooth filter to apply, and must be one of
6292 the following values: "blur", "blur_no_scale", "median", "gaussian",
6293 or "bilateral". The default value is "gaussian".
6295 The meaning of @var{param1}, @var{param2}, @var{param3}, and @var{param4}
6296 depend on the smooth type. @var{param1} and
6297 @var{param2} accept integer positive values or 0. @var{param3} and
6298 @var{param4} accept floating point values.
6300 The default value for @var{param1} is 3. The default value for the
6301 other parameters is 0.
6303 These parameters correspond to the parameters assigned to the
6304 libopencv function @code{cvSmooth}.
6309 Overlay one video on top of another.
6311 It takes two inputs and has one output. The first input is the "main"
6312 video on which the second input is overlayed.
6314 It accepts the following parameters:
6316 A description of the accepted options follows.
6321 Set the expression for the x and y coordinates of the overlayed video
6322 on the main video. Default value is "0" for both expressions. In case
6323 the expression is invalid, it is set to a huge value (meaning that the
6324 overlay will not be displayed within the output visible area).
6327 The action to take when EOF is encountered on the secondary input; it accepts
6328 one of the following values:
6332 Repeat the last frame (the default).
6336 Pass the main input through.
6340 Set when the expressions for @option{x}, and @option{y} are evaluated.
6342 It accepts the following values:
6345 only evaluate expressions once during the filter initialization or
6346 when a command is processed
6349 evaluate expressions for each incoming frame
6352 Default value is @samp{frame}.
6355 If set to 1, force the output to terminate when the shortest input
6356 terminates. Default value is 0.
6359 Set the format for the output video.
6361 It accepts the following values:
6376 Default value is @samp{yuv420}.
6378 @item rgb @emph{(deprecated)}
6379 If set to 1, force the filter to accept inputs in the RGB
6380 color space. Default value is 0. This option is deprecated, use
6381 @option{format} instead.
6384 If set to 1, force the filter to draw the last overlay frame over the
6385 main input until the end of the stream. A value of 0 disables this
6386 behavior. Default value is 1.
6389 The @option{x}, and @option{y} expressions can contain the following
6395 The main input width and height.
6399 The overlay input width and height.
6403 The computed values for @var{x} and @var{y}. They are evaluated for
6408 horizontal and vertical chroma subsample values of the output
6409 format. For example for the pixel format "yuv422p" @var{hsub} is 2 and
6413 the number of input frame, starting from 0
6416 the position in the file of the input frame, NAN if unknown
6419 The timestamp, expressed in seconds. It's NAN if the input timestamp is unknown.
6423 Note that the @var{n}, @var{pos}, @var{t} variables are available only
6424 when evaluation is done @emph{per frame}, and will evaluate to NAN
6425 when @option{eval} is set to @samp{init}.
6427 Be aware that frames are taken from each input video in timestamp
6428 order, hence, if their initial timestamps differ, it is a good idea
6429 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
6430 have them begin in the same zero timestamp, as the example for
6431 the @var{movie} filter does.
6433 You can chain together more overlays but you should test the
6434 efficiency of such approach.
6436 @subsection Commands
6438 This filter supports the following commands:
6442 Modify the x and y of the overlay input.
6443 The command accepts the same syntax of the corresponding option.
6445 If the specified expression is not valid, it is kept at its current
6449 @subsection Examples
6453 Draw the overlay at 10 pixels from the bottom right corner of the main
6456 overlay=main_w-overlay_w-10:main_h-overlay_h-10
6459 Using named options the example above becomes:
6461 overlay=x=main_w-overlay_w-10:y=main_h-overlay_h-10
6465 Insert a transparent PNG logo in the bottom left corner of the input,
6466 using the @command{ffmpeg} tool with the @code{-filter_complex} option:
6468 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
6472 Insert 2 different transparent PNG logos (second logo on bottom
6473 right corner) using the @command{ffmpeg} tool:
6475 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
6479 Add a transparent color layer on top of the main video; @code{WxH}
6480 must specify the size of the main input to the overlay filter:
6482 color=color=red@@.3:size=WxH [over]; [in][over] overlay [out]
6486 Play an original video and a filtered version (here with the deshake
6487 filter) side by side using the @command{ffplay} tool:
6489 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
6492 The above command is the same as:
6494 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
6498 Make a sliding overlay appearing from the left to the right top part of the
6499 screen starting since time 2:
6501 overlay=x='if(gte(t,2), -w+(t-2)*20, NAN)':y=0
6505 Compose output by putting two input videos side to side:
6507 ffmpeg -i left.avi -i right.avi -filter_complex "
6508 nullsrc=size=200x100 [background];
6509 [0:v] setpts=PTS-STARTPTS, scale=100x100 [left];
6510 [1:v] setpts=PTS-STARTPTS, scale=100x100 [right];
6511 [background][left] overlay=shortest=1 [background+left];
6512 [background+left][right] overlay=shortest=1:x=100 [left+right]
6517 Mask 10-20 seconds of a video by applying the delogo filter to a section
6519 ffmpeg -i test.avi -codec:v:0 wmv2 -ar 11025 -b:v 9000k
6520 -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]'
6525 Chain several overlays in cascade:
6527 nullsrc=s=200x200 [bg];
6528 testsrc=s=100x100, split=4 [in0][in1][in2][in3];
6529 [in0] lutrgb=r=0, [bg] overlay=0:0 [mid0];
6530 [in1] lutrgb=g=0, [mid0] overlay=100:0 [mid1];
6531 [in2] lutrgb=b=0, [mid1] overlay=0:100 [mid2];
6532 [in3] null, [mid2] overlay=100:100 [out0]
6539 Apply Overcomplete Wavelet denoiser.
6541 The filter accepts the following options:
6547 Larger depth values will denoise lower frequency components more, but
6548 slow down filtering.
6550 Must be an int in the range 8-16, default is @code{8}.
6552 @item luma_strength, ls
6555 Must be a double value in the range 0-1000, default is @code{1.0}.
6557 @item chroma_strength, cs
6558 Set chroma strength.
6560 Must be a double value in the range 0-1000, default is @code{1.0}.
6565 Add paddings to the input image, and place the original input at the
6566 provided @var{x}, @var{y} coordinates.
6568 It accepts the following parameters:
6573 Specify an expression for the size of the output image with the
6574 paddings added. If the value for @var{width} or @var{height} is 0, the
6575 corresponding input size is used for the output.
6577 The @var{width} expression can reference the value set by the
6578 @var{height} expression, and vice versa.
6580 The default value of @var{width} and @var{height} is 0.
6584 Specify the offsets to place the input image at within the padded area,
6585 with respect to the top/left border of the output image.
6587 The @var{x} expression can reference the value set by the @var{y}
6588 expression, and vice versa.
6590 The default value of @var{x} and @var{y} is 0.
6593 Specify the color of the padded area. For the syntax of this option,
6594 check the "Color" section in the ffmpeg-utils manual.
6596 The default value of @var{color} is "black".
6599 The value for the @var{width}, @var{height}, @var{x}, and @var{y}
6600 options are expressions containing the following constants:
6605 The input video width and height.
6609 These are the same as @var{in_w} and @var{in_h}.
6613 The output width and height (the size of the padded area), as
6614 specified by the @var{width} and @var{height} expressions.
6618 These are the same as @var{out_w} and @var{out_h}.
6622 The x and y offsets as specified by the @var{x} and @var{y}
6623 expressions, or NAN if not yet specified.
6626 same as @var{iw} / @var{ih}
6629 input sample aspect ratio
6632 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
6636 The horizontal and vertical chroma subsample values. For example for the
6637 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
6640 @subsection Examples
6644 Add paddings with the color "violet" to the input video. The output video
6645 size is 640x480, and the top-left corner of the input video is placed at
6648 pad=640:480:0:40:violet
6651 The example above is equivalent to the following command:
6653 pad=width=640:height=480:x=0:y=40:color=violet
6657 Pad the input to get an output with dimensions increased by 3/2,
6658 and put the input video at the center of the padded area:
6660 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
6664 Pad the input to get a squared output with size equal to the maximum
6665 value between the input width and height, and put the input video at
6666 the center of the padded area:
6668 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
6672 Pad the input to get a final w/h ratio of 16:9:
6674 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
6678 In case of anamorphic video, in order to set the output display aspect
6679 correctly, it is necessary to use @var{sar} in the expression,
6680 according to the relation:
6682 (ih * X / ih) * sar = output_dar
6683 X = output_dar / sar
6686 Thus the previous example needs to be modified to:
6688 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
6692 Double the output size and put the input video in the bottom-right
6693 corner of the output padded area:
6695 pad="2*iw:2*ih:ow-iw:oh-ih"
6699 @section perspective
6701 Correct perspective of video not recorded perpendicular to the screen.
6703 A description of the accepted parameters follows.
6714 Set coordinates expression for top left, top right, bottom left and bottom right corners.
6715 Default values are @code{0:0:W:0:0:H:W:H} with which perspective will remain unchanged.
6717 The expressions can use the following variables:
6722 the width and height of video frame.
6726 Set interpolation for perspective correction.
6728 It accepts the following values:
6734 Default value is @samp{linear}.
6739 Delay interlaced video by one field time so that the field order changes.
6741 The intended use is to fix PAL movies that have been captured with the
6742 opposite field order to the film-to-video transfer.
6744 A description of the accepted parameters follows.
6750 It accepts the following values:
6753 Capture field order top-first, transfer bottom-first.
6754 Filter will delay the bottom field.
6757 Capture field order bottom-first, transfer top-first.
6758 Filter will delay the top field.
6761 Capture and transfer with the same field order. This mode only exists
6762 for the documentation of the other options to refer to, but if you
6763 actually select it, the filter will faithfully do nothing.
6766 Capture field order determined automatically by field flags, transfer
6768 Filter selects among @samp{t} and @samp{b} modes on a frame by frame
6769 basis using field flags. If no field information is available,
6770 then this works just like @samp{u}.
6773 Capture unknown or varying, transfer opposite.
6774 Filter selects among @samp{t} and @samp{b} on a frame by frame basis by
6775 analyzing the images and selecting the alternative that produces best
6776 match between the fields.
6779 Capture top-first, transfer unknown or varying.
6780 Filter selects among @samp{t} and @samp{p} using image analysis.
6783 Capture bottom-first, transfer unknown or varying.
6784 Filter selects among @samp{b} and @samp{p} using image analysis.
6787 Capture determined by field flags, transfer unknown or varying.
6788 Filter selects among @samp{t}, @samp{b} and @samp{p} using field flags and
6789 image analysis. If no field information is available, then this works just
6790 like @samp{U}. This is the default mode.
6793 Both capture and transfer unknown or varying.
6794 Filter selects among @samp{t}, @samp{b} and @samp{p} using image analysis only.
6798 @section pixdesctest
6800 Pixel format descriptor test filter, mainly useful for internal
6801 testing. The output video should be equal to the input video.
6805 format=monow, pixdesctest
6808 can be used to test the monowhite pixel format descriptor definition.
6812 Enable the specified chain of postprocessing subfilters using libpostproc. This
6813 library should be automatically selected with a GPL build (@code{--enable-gpl}).
6814 Subfilters must be separated by '/' and can be disabled by prepending a '-'.
6815 Each subfilter and some options have a short and a long name that can be used
6816 interchangeably, i.e. dr/dering are the same.
6818 The filters accept the following options:
6822 Set postprocessing subfilters string.
6825 All subfilters share common options to determine their scope:
6829 Honor the quality commands for this subfilter.
6832 Do chrominance filtering, too (default).
6835 Do luminance filtering only (no chrominance).
6838 Do chrominance filtering only (no luminance).
6841 These options can be appended after the subfilter name, separated by a '|'.
6843 Available subfilters are:
6846 @item hb/hdeblock[|difference[|flatness]]
6847 Horizontal deblocking filter
6850 Difference factor where higher values mean more deblocking (default: @code{32}).
6852 Flatness threshold where lower values mean more deblocking (default: @code{39}).
6855 @item vb/vdeblock[|difference[|flatness]]
6856 Vertical deblocking filter
6859 Difference factor where higher values mean more deblocking (default: @code{32}).
6861 Flatness threshold where lower values mean more deblocking (default: @code{39}).
6864 @item ha/hadeblock[|difference[|flatness]]
6865 Accurate horizontal deblocking filter
6868 Difference factor where higher values mean more deblocking (default: @code{32}).
6870 Flatness threshold where lower values mean more deblocking (default: @code{39}).
6873 @item va/vadeblock[|difference[|flatness]]
6874 Accurate vertical deblocking filter
6877 Difference factor where higher values mean more deblocking (default: @code{32}).
6879 Flatness threshold where lower values mean more deblocking (default: @code{39}).
6883 The horizontal and vertical deblocking filters share the difference and
6884 flatness values so you cannot set different horizontal and vertical
6889 Experimental horizontal deblocking filter
6892 Experimental vertical deblocking filter
6897 @item tn/tmpnoise[|threshold1[|threshold2[|threshold3]]], temporal noise reducer
6900 larger -> stronger filtering
6902 larger -> stronger filtering
6904 larger -> stronger filtering
6907 @item al/autolevels[:f/fullyrange], automatic brightness / contrast correction
6910 Stretch luminance to @code{0-255}.
6913 @item lb/linblenddeint
6914 Linear blend deinterlacing filter that deinterlaces the given block by
6915 filtering all lines with a @code{(1 2 1)} filter.
6917 @item li/linipoldeint
6918 Linear interpolating deinterlacing filter that deinterlaces the given block by
6919 linearly interpolating every second line.
6921 @item ci/cubicipoldeint
6922 Cubic interpolating deinterlacing filter deinterlaces the given block by
6923 cubically interpolating every second line.
6925 @item md/mediandeint
6926 Median deinterlacing filter that deinterlaces the given block by applying a
6927 median filter to every second line.
6929 @item fd/ffmpegdeint
6930 FFmpeg deinterlacing filter that deinterlaces the given block by filtering every
6931 second line with a @code{(-1 4 2 4 -1)} filter.
6934 Vertically applied FIR lowpass deinterlacing filter that deinterlaces the given
6935 block by filtering all lines with a @code{(-1 2 6 2 -1)} filter.
6937 @item fq/forceQuant[|quantizer]
6938 Overrides the quantizer table from the input with the constant quantizer you
6946 Default pp filter combination (@code{hb|a,vb|a,dr|a})
6949 Fast pp filter combination (@code{h1|a,v1|a,dr|a})
6952 High quality pp filter combination (@code{ha|a|128|7,va|a,dr|a})
6955 @subsection Examples
6959 Apply horizontal and vertical deblocking, deringing and automatic
6960 brightness/contrast:
6966 Apply default filters without brightness/contrast correction:
6972 Apply default filters and temporal denoiser:
6974 pp=default/tmpnoise|1|2|3
6978 Apply deblocking on luminance only, and switch vertical deblocking on or off
6979 automatically depending on available CPU time:
6987 Obtain the average, maximum and minimum PSNR (Peak Signal to Noise
6988 Ratio) between two input videos.
6990 This filter takes in input two input videos, the first input is
6991 considered the "main" source and is passed unchanged to the
6992 output. The second input is used as a "reference" video for computing
6995 Both video inputs must have the same resolution and pixel format for
6996 this filter to work correctly. Also it assumes that both inputs
6997 have the same number of frames, which are compared one by one.
6999 The obtained average PSNR is printed through the logging system.
7001 The filter stores the accumulated MSE (mean squared error) of each
7002 frame, and at the end of the processing it is averaged across all frames
7003 equally, and the following formula is applied to obtain the PSNR:
7006 PSNR = 10*log10(MAX^2/MSE)
7009 Where MAX is the average of the maximum values of each component of the
7012 The description of the accepted parameters follows.
7016 If specified the filter will use the named file to save the PSNR of
7017 each individual frame.
7020 The file printed if @var{stats_file} is selected, contains a sequence of
7021 key/value pairs of the form @var{key}:@var{value} for each compared
7024 A description of each shown parameter follows:
7028 sequential number of the input frame, starting from 1
7031 Mean Square Error pixel-by-pixel average difference of the compared
7032 frames, averaged over all the image components.
7034 @item mse_y, mse_u, mse_v, mse_r, mse_g, mse_g, mse_a
7035 Mean Square Error pixel-by-pixel average difference of the compared
7036 frames for the component specified by the suffix.
7038 @item psnr_y, psnr_u, psnr_v, psnr_r, psnr_g, psnr_b, psnr_a
7039 Peak Signal to Noise ratio of the compared frames for the component
7040 specified by the suffix.
7045 movie=ref_movie.mpg, setpts=PTS-STARTPTS [main];
7046 [main][ref] psnr="stats_file=stats.log" [out]
7049 On this example the input file being processed is compared with the
7050 reference file @file{ref_movie.mpg}. The PSNR of each individual frame
7051 is stored in @file{stats.log}.
7056 Pulldown reversal (inverse telecine) filter, capable of handling mixed
7057 hard-telecine, 24000/1001 fps progressive, and 30000/1001 fps progressive
7060 The pullup filter is designed to take advantage of future context in making
7061 its decisions. This filter is stateless in the sense that it does not lock
7062 onto a pattern to follow, but it instead looks forward to the following
7063 fields in order to identify matches and rebuild progressive frames.
7065 To produce content with an even framerate, insert the fps filter after
7066 pullup, use @code{fps=24000/1001} if the input frame rate is 29.97fps,
7067 @code{fps=24} for 30fps and the (rare) telecined 25fps input.
7069 The filter accepts the following options:
7076 These options set the amount of "junk" to ignore at the left, right, top, and
7077 bottom of the image, respectively. Left and right are in units of 8 pixels,
7078 while top and bottom are in units of 2 lines.
7079 The default is 8 pixels on each side.
7082 Set the strict breaks. Setting this option to 1 will reduce the chances of
7083 filter generating an occasional mismatched frame, but it may also cause an
7084 excessive number of frames to be dropped during high motion sequences.
7085 Conversely, setting it to -1 will make filter match fields more easily.
7086 This may help processing of video where there is slight blurring between
7087 the fields, but may also cause there to be interlaced frames in the output.
7088 Default value is @code{0}.
7091 Set the metric plane to use. It accepts the following values:
7097 Use chroma blue plane.
7100 Use chroma red plane.
7103 This option may be set to use chroma plane instead of the default luma plane
7104 for doing filter's computations. This may improve accuracy on very clean
7105 source material, but more likely will decrease accuracy, especially if there
7106 is chroma noise (rainbow effect) or any grayscale video.
7107 The main purpose of setting @option{mp} to a chroma plane is to reduce CPU
7108 load and make pullup usable in realtime on slow machines.
7111 For best results (without duplicated frames in the output file) it is
7112 necessary to change the output frame rate. For example, to inverse
7113 telecine NTSC input:
7115 ffmpeg -i input -vf pullup -r 24000/1001 ...
7120 Suppress a TV station logo, using an image file to determine which
7121 pixels comprise the logo. It works by filling in the pixels that
7122 comprise the logo with neighboring pixels.
7124 The filter accepts the following options:
7128 Set the filter bitmap file, which can be any image format supported by
7129 libavformat. The width and height of the image file must match those of the
7130 video stream being processed.
7133 Pixels in the provided bitmap image with a value of zero are not
7134 considered part of the logo, non-zero pixels are considered part of
7135 the logo. If you use white (255) for the logo and black (0) for the
7136 rest, you will be safe. For making the filter bitmap, it is
7137 recommended to take a screen capture of a black frame with the logo
7138 visible, and then using a threshold filter followed by the erode
7139 filter once or twice.
7141 If needed, little splotches can be fixed manually. Remember that if
7142 logo pixels are not covered, the filter quality will be much
7143 reduced. Marking too many pixels as part of the logo does not hurt as
7144 much, but it will increase the amount of blurring needed to cover over
7145 the image and will destroy more information than necessary, and extra
7146 pixels will slow things down on a large logo.
7150 Rotate video by an arbitrary angle expressed in radians.
7152 The filter accepts the following options:
7154 A description of the optional parameters follows.
7157 Set an expression for the angle by which to rotate the input video
7158 clockwise, expressed as a number of radians. A negative value will
7159 result in a counter-clockwise rotation. By default it is set to "0".
7161 This expression is evaluated for each frame.
7164 Set the output width expression, default value is "iw".
7165 This expression is evaluated just once during configuration.
7168 Set the output height expression, default value is "ih".
7169 This expression is evaluated just once during configuration.
7172 Enable bilinear interpolation if set to 1, a value of 0 disables
7173 it. Default value is 1.
7176 Set the color used to fill the output area not covered by the rotated
7177 image. For the generalsyntax of this option, check the "Color" section in the
7178 ffmpeg-utils manual. If the special value "none" is selected then no
7179 background is printed (useful for example if the background is never shown).
7181 Default value is "black".
7184 The expressions for the angle and the output size can contain the
7185 following constants and functions:
7189 sequential number of the input frame, starting from 0. It is always NAN
7190 before the first frame is filtered.
7193 time in seconds of the input frame, it is set to 0 when the filter is
7194 configured. It is always NAN before the first frame is filtered.
7198 horizontal and vertical chroma subsample values. For example for the
7199 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
7203 the input video width and height
7207 the output width and height, that is the size of the padded area as
7208 specified by the @var{width} and @var{height} expressions
7212 the minimal width/height required for completely containing the input
7213 video rotated by @var{a} radians.
7215 These are only available when computing the @option{out_w} and
7216 @option{out_h} expressions.
7219 @subsection Examples
7223 Rotate the input by PI/6 radians clockwise:
7229 Rotate the input by PI/6 radians counter-clockwise:
7235 Rotate the input by 45 degrees clockwise:
7241 Apply a constant rotation with period T, starting from an angle of PI/3:
7243 rotate=PI/3+2*PI*t/T
7247 Make the input video rotation oscillating with a period of T
7248 seconds and an amplitude of A radians:
7250 rotate=A*sin(2*PI/T*t)
7254 Rotate the video, output size is chosen so that the whole rotating
7255 input video is always completely contained in the output:
7257 rotate='2*PI*t:ow=hypot(iw,ih):oh=ow'
7261 Rotate the video, reduce the output size so that no background is ever
7264 rotate=2*PI*t:ow='min(iw,ih)/sqrt(2)':oh=ow:c=none
7268 @subsection Commands
7270 The filter supports the following commands:
7274 Set the angle expression.
7275 The command accepts the same syntax of the corresponding option.
7277 If the specified expression is not valid, it is kept at its current
7283 Apply Shape Adaptive Blur.
7285 The filter accepts the following options:
7288 @item luma_radius, lr
7289 Set luma blur filter strength, must be a value in range 0.1-4.0, default
7290 value is 1.0. A greater value will result in a more blurred image, and
7291 in slower processing.
7293 @item luma_pre_filter_radius, lpfr
7294 Set luma pre-filter radius, must be a value in the 0.1-2.0 range, default
7297 @item luma_strength, ls
7298 Set luma maximum difference between pixels to still be considered, must
7299 be a value in the 0.1-100.0 range, default value is 1.0.
7301 @item chroma_radius, cr
7302 Set chroma blur filter strength, must be a value in range 0.1-4.0. A
7303 greater value will result in a more blurred image, and in slower
7306 @item chroma_pre_filter_radius, cpfr
7307 Set chroma pre-filter radius, must be a value in the 0.1-2.0 range.
7309 @item chroma_strength, cs
7310 Set chroma maximum difference between pixels to still be considered,
7311 must be a value in the 0.1-100.0 range.
7314 Each chroma option value, if not explicitly specified, is set to the
7315 corresponding luma option value.
7320 Scale (resize) the input video, using the libswscale library.
7322 The scale filter forces the output display aspect ratio to be the same
7323 of the input, by changing the output sample aspect ratio.
7325 If the input image format is different from the format requested by
7326 the next filter, the scale filter will convert the input to the
7330 The filter accepts the following options, or any of the options
7331 supported by the libswscale scaler.
7333 See @ref{scaler_options,,the ffmpeg-scaler manual,ffmpeg-scaler} for
7334 the complete list of scaler options.
7339 Set the output video dimension expression. Default value is the input
7342 If the value is 0, the input width is used for the output.
7344 If one of the values is -1, the scale filter will use a value that
7345 maintains the aspect ratio of the input image, calculated from the
7346 other specified dimension. If both of them are -1, the input size is
7349 If one of the values is -n with n > 1, the scale filter will also use a value
7350 that maintains the aspect ratio of the input image, calculated from the other
7351 specified dimension. After that it will, however, make sure that the calculated
7352 dimension is divisible by n and adjust the value if necessary.
7354 See below for the list of accepted constants for use in the dimension
7358 Set the interlacing mode. It accepts the following values:
7362 Force interlaced aware scaling.
7365 Do not apply interlaced scaling.
7368 Select interlaced aware scaling depending on whether the source frames
7369 are flagged as interlaced or not.
7372 Default value is @samp{0}.
7375 Set libswscale scaling flags. See
7376 @ref{sws_flags,,the ffmpeg-scaler manual,ffmpeg-scaler} for the
7377 complete list of values. If not explicitly specified the filter applies
7381 Set the video size. For the syntax of this option, check the "Video size"
7382 section in the ffmpeg-utils manual.
7384 @item in_color_matrix
7385 @item out_color_matrix
7386 Set in/output YCbCr color space type.
7388 This allows the autodetected value to be overridden as well as allows forcing
7389 a specific value used for the output and encoder.
7391 If not specified, the color space type depends on the pixel format.
7397 Choose automatically.
7400 Format conforming to International Telecommunication Union (ITU)
7401 Recommendation BT.709.
7404 Set color space conforming to the United States Federal Communications
7405 Commission (FCC) Code of Federal Regulations (CFR) Title 47 (2003) 73.682 (a).
7408 Set color space conforming to:
7412 ITU Radiocommunication Sector (ITU-R) Recommendation BT.601
7415 ITU-R Rec. BT.470-6 (1998) Systems B, B1, and G
7418 Society of Motion Picture and Television Engineers (SMPTE) ST 170:2004
7423 Set color space conforming to SMPTE ST 240:1999.
7428 Set in/output YCbCr sample range.
7430 This allows the autodetected value to be overridden as well as allows forcing
7431 a specific value used for the output and encoder. If not specified, the
7432 range depends on the pixel format. Possible values:
7436 Choose automatically.
7439 Set full range (0-255 in case of 8-bit luma).
7442 Set "MPEG" range (16-235 in case of 8-bit luma).
7445 @item force_original_aspect_ratio
7446 Enable decreasing or increasing output video width or height if necessary to
7447 keep the original aspect ratio. Possible values:
7451 Scale the video as specified and disable this feature.
7454 The output video dimensions will automatically be decreased if needed.
7457 The output video dimensions will automatically be increased if needed.
7461 One useful instance of this option is that when you know a specific device's
7462 maximum allowed resolution, you can use this to limit the output video to
7463 that, while retaining the aspect ratio. For example, device A allows
7464 1280x720 playback, and your video is 1920x800. Using this option (set it to
7465 decrease) and specifying 1280x720 to the command line makes the output
7468 Please note that this is a different thing than specifying -1 for @option{w}
7469 or @option{h}, you still need to specify the output resolution for this option
7474 The values of the @option{w} and @option{h} options are expressions
7475 containing the following constants:
7480 The input width and height
7484 These are the same as @var{in_w} and @var{in_h}.
7488 The output (scaled) width and height
7492 These are the same as @var{out_w} and @var{out_h}
7495 The same as @var{iw} / @var{ih}
7498 input sample aspect ratio
7501 The input display aspect ratio. Calculated from @code{(iw / ih) * sar}.
7505 horizontal and vertical input chroma subsample values. For example for the
7506 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
7510 horizontal and vertical output chroma subsample values. For example for the
7511 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
7514 @subsection Examples
7518 Scale the input video to a size of 200x100
7523 This is equivalent to:
7534 Specify a size abbreviation for the output size:
7539 which can also be written as:
7545 Scale the input to 2x:
7551 The above is the same as:
7557 Scale the input to 2x with forced interlaced scaling:
7559 scale=2*iw:2*ih:interl=1
7563 Scale the input to half size:
7569 Increase the width, and set the height to the same size:
7582 Increase the height, and set the width to 3/2 of the height:
7584 scale=w=3/2*oh:h=3/5*ih
7588 Increase the size, making the size a multiple of the chroma
7591 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
7595 Increase the width to a maximum of 500 pixels,
7596 keeping the same aspect ratio as the input:
7598 scale=w='min(500\, iw*3/2):h=-1'
7602 @section separatefields
7604 The @code{separatefields} takes a frame-based video input and splits
7605 each frame into its components fields, producing a new half height clip
7606 with twice the frame rate and twice the frame count.
7608 This filter use field-dominance information in frame to decide which
7609 of each pair of fields to place first in the output.
7610 If it gets it wrong use @ref{setfield} filter before @code{separatefields} filter.
7612 @section setdar, setsar
7614 The @code{setdar} filter sets the Display Aspect Ratio for the filter
7617 This is done by changing the specified Sample (aka Pixel) Aspect
7618 Ratio, according to the following equation:
7620 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
7623 Keep in mind that the @code{setdar} filter does not modify the pixel
7624 dimensions of the video frame. Also, the display aspect ratio set by
7625 this filter may be changed by later filters in the filterchain,
7626 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
7629 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
7630 the filter output video.
7632 Note that as a consequence of the application of this filter, the
7633 output display aspect ratio will change according to the equation
7636 Keep in mind that the sample aspect ratio set by the @code{setsar}
7637 filter may be changed by later filters in the filterchain, e.g. if
7638 another "setsar" or a "setdar" filter is applied.
7640 It accepts the following parameters:
7643 @item r, ratio, dar (@code{setdar} only), sar (@code{setsar} only)
7644 Set the aspect ratio used by the filter.
7646 The parameter can be a floating point number string, an expression, or
7647 a string of the form @var{num}:@var{den}, where @var{num} and
7648 @var{den} are the numerator and denominator of the aspect ratio. If
7649 the parameter is not specified, it is assumed the value "0".
7650 In case the form "@var{num}:@var{den}" is used, the @code{:} character
7654 Set the maximum integer value to use for expressing numerator and
7655 denominator when reducing the expressed aspect ratio to a rational.
7656 Default value is @code{100}.
7660 The parameter @var{sar} is an expression containing
7661 the following constants:
7665 These are approximated values for the mathematical constants e
7666 (Euler's number), pi (Greek pi), and phi (the golden ratio).
7669 The input width and height.
7672 These are the same as @var{w} / @var{h}.
7675 The input sample aspect ratio.
7678 The input display aspect ratio. It is the same as
7679 (@var{w} / @var{h}) * @var{sar}.
7682 Horizontal and vertical chroma subsample values. For example, for the
7683 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
7686 @subsection Examples
7691 To change the display aspect ratio to 16:9, specify one of the following:
7699 To change the sample aspect ratio to 10:11, specify:
7705 To set a display aspect ratio of 16:9, and specify a maximum integer value of
7706 1000 in the aspect ratio reduction, use the command:
7708 setdar=ratio=16/9:max=1000
7716 Force field for the output video frame.
7718 The @code{setfield} filter marks the interlace type field for the
7719 output frames. It does not change the input frame, but only sets the
7720 corresponding property, which affects how the frame is treated by
7721 following filters (e.g. @code{fieldorder} or @code{yadif}).
7723 The filter accepts the following options:
7728 Available values are:
7732 Keep the same field property.
7735 Mark the frame as bottom-field-first.
7738 Mark the frame as top-field-first.
7741 Mark the frame as progressive.
7747 Show a line containing various information for each input video frame.
7748 The input video is not modified.
7750 The shown line contains a sequence of key/value pairs of the form
7751 @var{key}:@var{value}.
7753 The following values are shown in the output:
7757 The (sequential) number of the input frame, starting from 0.
7760 The Presentation TimeStamp of the input frame, expressed as a number of
7761 time base units. The time base unit depends on the filter input pad.
7764 The Presentation TimeStamp of the input frame, expressed as a number of
7768 The position of the frame in the input stream, or -1 if this information is
7769 unavailable and/or meaningless (for example in case of synthetic video).
7772 The pixel format name.
7775 The sample aspect ratio of the input frame, expressed in the form
7776 @var{num}/@var{den}.
7779 The size of the input frame. For the syntax of this option, check the "Video size"
7780 section in the ffmpeg-utils manual.
7783 The type of interlaced mode ("P" for "progressive", "T" for top field first, "B"
7784 for bottom field first).
7787 This is 1 if the frame is a key frame, 0 otherwise.
7790 The picture type of the input frame ("I" for an I-frame, "P" for a
7791 P-frame, "B" for a B-frame, or "?" for an unknown type).
7792 Also refer to the documentation of the @code{AVPictureType} enum and of
7793 the @code{av_get_picture_type_char} function defined in
7794 @file{libavutil/avutil.h}.
7797 The Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame.
7799 @item plane_checksum
7800 The Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
7801 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]".
7804 @section shuffleplanes
7806 Reorder and/or duplicate video planes.
7808 It accepts the following parameters:
7813 The index of the input plane to be used as the first output plane.
7816 The index of the input plane to be used as the second output plane.
7819 The index of the input plane to be used as the third output plane.
7822 The index of the input plane to be used as the fourth output plane.
7826 The first plane has the index 0. The default is to keep the input unchanged.
7828 Swap the second and third planes of the input:
7830 ffmpeg -i INPUT -vf shuffleplanes=0:2:1:3 OUTPUT
7833 @section signalstats
7834 Evaluate various visual metrics that assist in determining issues associated
7835 with the digitization of analog video media.
7837 By default the filter will log these metadata values:
7841 Display the minimal Y value contained within the input frame. Expressed in
7845 Display the Y value at the 10% percentile within the input frame. Expressed in
7849 Display the average Y value within the input frame. Expressed in range of
7853 Display the Y value at the 90% percentile within the input frame. Expressed in
7857 Display the maximum Y value contained within the input frame. Expressed in
7861 Display the minimal U value contained within the input frame. Expressed in
7865 Display the U value at the 10% percentile within the input frame. Expressed in
7869 Display the average U value within the input frame. Expressed in range of
7873 Display the U value at the 90% percentile within the input frame. Expressed in
7877 Display the maximum U value contained within the input frame. Expressed in
7881 Display the minimal V value contained within the input frame. Expressed in
7885 Display the V value at the 10% percentile within the input frame. Expressed in
7889 Display the average V value within the input frame. Expressed in range of
7893 Display the V value at the 90% percentile within the input frame. Expressed in
7897 Display the maximum V value contained within the input frame. Expressed in
7901 Display the minimal saturation value contained within the input frame.
7902 Expressed in range of [0-~181.02].
7905 Display the saturation value at the 10% percentile within the input frame.
7906 Expressed in range of [0-~181.02].
7909 Display the average saturation value within the input frame. Expressed in range
7913 Display the saturation value at the 90% percentile within the input frame.
7914 Expressed in range of [0-~181.02].
7917 Display the maximum saturation value contained within the input frame.
7918 Expressed in range of [0-~181.02].
7921 Display the median value for hue within the input frame. Expressed in range of
7925 Display the average value for hue within the input frame. Expressed in range of
7929 Display the average of sample value difference between all values of the Y
7930 plane in the current frame and corresponding values of the previous input frame.
7931 Expressed in range of [0-255].
7934 Display the average of sample value difference between all values of the U
7935 plane in the current frame and corresponding values of the previous input frame.
7936 Expressed in range of [0-255].
7939 Display the average of sample value difference between all values of the V
7940 plane in the current frame and corresponding values of the previous input frame.
7941 Expressed in range of [0-255].
7944 The filter accepts the following options:
7950 @option{stat} specify an additional form of image analysis.
7951 @option{out} output video with the specified type of pixel highlighted.
7953 Both options accept the following values:
7957 Identify @var{temporal outliers} pixels. A @var{temporal outlier} is a pixel
7958 unlike the neighboring pixels of the same field. Examples of temporal outliers
7959 include the results of video dropouts, head clogs, or tape tracking issues.
7962 Identify @var{vertical line repetition}. Vertical line repetition includes
7963 similar rows of pixels within a frame. In born-digital video vertical line
7964 repetition is common, but this pattern is uncommon in video digitized from an
7965 analog source. When it occurs in video that results from the digitization of an
7966 analog source it can indicate concealment from a dropout compensator.
7969 Identify pixels that fall outside of legal broadcast range.
7973 Set the highlight color for the @option{out} option. The default color is
7977 @subsection Examples
7981 Output data of various video metrics:
7983 ffprobe -f lavfi movie=example.mov,signalstats="stat=tout+vrep+brng" -show_frames
7987 Output specific data about the minimum and maximum values of the Y plane per frame:
7989 ffprobe -f lavfi movie=example.mov,signalstats -show_entries frame_tags=lavfi.signalstats.YMAX,lavfi.signalstats.YMIN
7993 Playback video while highlighting pixels that are outside of broadcast range in red.
7995 ffplay example.mov -vf signalstats="out=brng:color=red"
7999 Playback video with signalstats metadata drawn over the frame.
8001 ffplay example.mov -vf signalstats=stat=brng+vrep+tout,drawtext=fontfile=FreeSerif.ttf:textfile=signalstat_drawtext.txt
8004 The contents of signalstat_drawtext.txt used in the command are:
8007 Y (%@{metadata:lavfi.signalstats.YMIN@}-%@{metadata:lavfi.signalstats.YMAX@})
8008 U (%@{metadata:lavfi.signalstats.UMIN@}-%@{metadata:lavfi.signalstats.UMAX@})
8009 V (%@{metadata:lavfi.signalstats.VMIN@}-%@{metadata:lavfi.signalstats.VMAX@})
8010 saturation maximum: %@{metadata:lavfi.signalstats.SATMAX@}
8018 Blur the input video without impacting the outlines.
8020 It accepts the following options:
8023 @item luma_radius, lr
8024 Set the luma radius. The option value must be a float number in
8025 the range [0.1,5.0] that specifies the variance of the gaussian filter
8026 used to blur the image (slower if larger). Default value is 1.0.
8028 @item luma_strength, ls
8029 Set the luma strength. The option value must be a float number
8030 in the range [-1.0,1.0] that configures the blurring. A value included
8031 in [0.0,1.0] will blur the image whereas a value included in
8032 [-1.0,0.0] will sharpen the image. Default value is 1.0.
8034 @item luma_threshold, lt
8035 Set the luma threshold used as a coefficient to determine
8036 whether a pixel should be blurred or not. The option value must be an
8037 integer in the range [-30,30]. A value of 0 will filter all the image,
8038 a value included in [0,30] will filter flat areas and a value included
8039 in [-30,0] will filter edges. Default value is 0.
8041 @item chroma_radius, cr
8042 Set the chroma radius. The option value must be a float number in
8043 the range [0.1,5.0] that specifies the variance of the gaussian filter
8044 used to blur the image (slower if larger). Default value is 1.0.
8046 @item chroma_strength, cs
8047 Set the chroma strength. The option value must be a float number
8048 in the range [-1.0,1.0] that configures the blurring. A value included
8049 in [0.0,1.0] will blur the image whereas a value included in
8050 [-1.0,0.0] will sharpen the image. Default value is 1.0.
8052 @item chroma_threshold, ct
8053 Set the chroma threshold used as a coefficient to determine
8054 whether a pixel should be blurred or not. The option value must be an
8055 integer in the range [-30,30]. A value of 0 will filter all the image,
8056 a value included in [0,30] will filter flat areas and a value included
8057 in [-30,0] will filter edges. Default value is 0.
8060 If a chroma option is not explicitly set, the corresponding luma value
8065 Convert between different stereoscopic image formats.
8067 The filters accept the following options:
8071 Set stereoscopic image format of input.
8073 Available values for input image formats are:
8076 side by side parallel (left eye left, right eye right)
8079 side by side crosseye (right eye left, left eye right)
8082 side by side parallel with half width resolution
8083 (left eye left, right eye right)
8086 side by side crosseye with half width resolution
8087 (right eye left, left eye right)
8090 above-below (left eye above, right eye below)
8093 above-below (right eye above, left eye below)
8096 above-below with half height resolution
8097 (left eye above, right eye below)
8100 above-below with half height resolution
8101 (right eye above, left eye below)
8104 alternating frames (left eye first, right eye second)
8107 alternating frames (right eye first, left eye second)
8109 Default value is @samp{sbsl}.
8113 Set stereoscopic image format of output.
8115 Available values for output image formats are all the input formats as well as:
8118 anaglyph red/blue gray
8119 (red filter on left eye, blue filter on right eye)
8122 anaglyph red/green gray
8123 (red filter on left eye, green filter on right eye)
8126 anaglyph red/cyan gray
8127 (red filter on left eye, cyan filter on right eye)
8130 anaglyph red/cyan half colored
8131 (red filter on left eye, cyan filter on right eye)
8134 anaglyph red/cyan color
8135 (red filter on left eye, cyan filter on right eye)
8138 anaglyph red/cyan color optimized with the least squares projection of dubois
8139 (red filter on left eye, cyan filter on right eye)
8142 anaglyph green/magenta gray
8143 (green filter on left eye, magenta filter on right eye)
8146 anaglyph green/magenta half colored
8147 (green filter on left eye, magenta filter on right eye)
8150 anaglyph green/magenta colored
8151 (green filter on left eye, magenta filter on right eye)
8154 anaglyph green/magenta color optimized with the least squares projection of dubois
8155 (green filter on left eye, magenta filter on right eye)
8158 anaglyph yellow/blue gray
8159 (yellow filter on left eye, blue filter on right eye)
8162 anaglyph yellow/blue half colored
8163 (yellow filter on left eye, blue filter on right eye)
8166 anaglyph yellow/blue colored
8167 (yellow filter on left eye, blue filter on right eye)
8170 anaglyph yellow/blue color optimized with the least squares projection of dubois
8171 (yellow filter on left eye, blue filter on right eye)
8174 interleaved rows (left eye has top row, right eye starts on next row)
8177 interleaved rows (right eye has top row, left eye starts on next row)
8180 mono output (left eye only)
8183 mono output (right eye only)
8186 Default value is @samp{arcd}.
8189 @subsection Examples
8193 Convert input video from side by side parallel to anaglyph yellow/blue dubois:
8199 Convert input video from above bellow (left eye above, right eye below) to side by side crosseye.
8207 Apply a simple postprocessing filter that compresses and decompresses the image
8208 at several (or - in the case of @option{quality} level @code{6} - all) shifts
8209 and average the results.
8211 The filter accepts the following options:
8215 Set quality. This option defines the number of levels for averaging. It accepts
8216 an integer in the range 0-6. If set to @code{0}, the filter will have no
8217 effect. A value of @code{6} means the higher quality. For each increment of
8218 that value the speed drops by a factor of approximately 2. Default value is
8222 Force a constant quantization parameter. If not set, the filter will use the QP
8223 from the video stream (if available).
8226 Set thresholding mode. Available modes are:
8230 Set hard thresholding (default).
8232 Set soft thresholding (better de-ringing effect, but likely blurrier).
8236 Enable the use of the QP from the B-Frames if set to @code{1}. Using this
8237 option may cause flicker since the B-Frames have often larger QP. Default is
8238 @code{0} (not enabled).
8244 Draw subtitles on top of input video using the libass library.
8246 To enable compilation of this filter you need to configure FFmpeg with
8247 @code{--enable-libass}. This filter also requires a build with libavcodec and
8248 libavformat to convert the passed subtitles file to ASS (Advanced Substation
8249 Alpha) subtitles format.
8251 The filter accepts the following options:
8255 Set the filename of the subtitle file to read. It must be specified.
8258 Specify the size of the original video, the video for which the ASS file
8259 was composed. For the syntax of this option, check the "Video size" section in
8260 the ffmpeg-utils manual. Due to a misdesign in ASS aspect ratio arithmetic,
8261 this is necessary to correctly scale the fonts if the aspect ratio has been
8265 Set subtitles input character encoding. @code{subtitles} filter only. Only
8266 useful if not UTF-8.
8268 @item stream_index, si
8269 Set subtitles stream index. @code{subtitles} filter only.
8272 If the first key is not specified, it is assumed that the first value
8273 specifies the @option{filename}.
8275 For example, to render the file @file{sub.srt} on top of the input
8276 video, use the command:
8281 which is equivalent to:
8283 subtitles=filename=sub.srt
8286 To render the default subtitles stream from file @file{video.mkv}, use:
8291 To render the second subtitles stream from that file, use:
8293 subtitles=video.mkv:si=1
8298 Scale the input by 2x and smooth using the Super2xSaI (Scale and
8299 Interpolate) pixel art scaling algorithm.
8301 Useful for enlarging pixel art images without reducing sharpness.
8308 Apply telecine process to the video.
8310 This filter accepts the following options:
8319 The default value is @code{top}.
8323 A string of numbers representing the pulldown pattern you wish to apply.
8324 The default value is @code{23}.
8328 Some typical patterns:
8333 24p: 2332 (preferred)
8340 24p: 222222222223 ("Euro pulldown")
8346 Select the most representative frame in a given sequence of consecutive frames.
8348 The filter accepts the following options:
8352 Set the frames batch size to analyze; in a set of @var{n} frames, the filter
8353 will pick one of them, and then handle the next batch of @var{n} frames until
8354 the end. Default is @code{100}.
8357 Since the filter keeps track of the whole frames sequence, a bigger @var{n}
8358 value will result in a higher memory usage, so a high value is not recommended.
8360 @subsection Examples
8364 Extract one picture each 50 frames:
8370 Complete example of a thumbnail creation with @command{ffmpeg}:
8372 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
8378 Tile several successive frames together.
8380 The filter accepts the following options:
8385 Set the grid size (i.e. the number of lines and columns). For the syntax of
8386 this option, check the "Video size" section in the ffmpeg-utils manual.
8389 Set the maximum number of frames to render in the given area. It must be less
8390 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
8391 the area will be used.
8394 Set the outer border margin in pixels.
8397 Set the inner border thickness (i.e. the number of pixels between frames). For
8398 more advanced padding options (such as having different values for the edges),
8399 refer to the pad video filter.
8402 Specify the color of the unused areaFor the syntax of this option, check the
8403 "Color" section in the ffmpeg-utils manual. The default value of @var{color}
8407 @subsection Examples
8411 Produce 8x8 PNG tiles of all keyframes (@option{-skip_frame nokey}) in a movie:
8413 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
8415 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
8416 duplicating each output frame to accommodate the originally detected frame
8420 Display @code{5} pictures in an area of @code{3x2} frames,
8421 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
8422 mixed flat and named options:
8424 tile=3x2:nb_frames=5:padding=7:margin=2
8430 Perform various types of temporal field interlacing.
8432 Frames are counted starting from 1, so the first input frame is
8435 The filter accepts the following options:
8440 Specify the mode of the interlacing. This option can also be specified
8441 as a value alone. See below for a list of values for this option.
8443 Available values are:
8447 Move odd frames into the upper field, even into the lower field,
8448 generating a double height frame at half frame rate.
8451 Only output even frames, odd frames are dropped, generating a frame with
8452 unchanged height at half frame rate.
8455 Only output odd frames, even frames are dropped, generating a frame with
8456 unchanged height at half frame rate.
8459 Expand each frame to full height, but pad alternate lines with black,
8460 generating a frame with double height at the same input frame rate.
8462 @item interleave_top, 4
8463 Interleave the upper field from odd frames with the lower field from
8464 even frames, generating a frame with unchanged height at half frame rate.
8466 @item interleave_bottom, 5
8467 Interleave the lower field from odd frames with the upper field from
8468 even frames, generating a frame with unchanged height at half frame rate.
8470 @item interlacex2, 6
8471 Double frame rate with unchanged height. Frames are inserted each
8472 containing the second temporal field from the previous input frame and
8473 the first temporal field from the next input frame. This mode relies on
8474 the top_field_first flag. Useful for interlaced video displays with no
8475 field synchronisation.
8478 Numeric values are deprecated but are accepted for backward
8479 compatibility reasons.
8481 Default mode is @code{merge}.
8484 Specify flags influencing the filter process.
8486 Available value for @var{flags} is:
8489 @item low_pass_filter, vlfp
8490 Enable vertical low-pass filtering in the filter.
8491 Vertical low-pass filtering is required when creating an interlaced
8492 destination from a progressive source which contains high-frequency
8493 vertical detail. Filtering will reduce interlace 'twitter' and Moire
8496 Vertical low-pass filtering can only be enabled for @option{mode}
8497 @var{interleave_top} and @var{interleave_bottom}.
8504 Transpose rows with columns in the input video and optionally flip it.
8506 It accepts the following parameters:
8511 Specify the transposition direction.
8513 Can assume the following values:
8515 @item 0, 4, cclock_flip
8516 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
8524 Rotate by 90 degrees clockwise, that is:
8532 Rotate by 90 degrees counterclockwise, that is:
8539 @item 3, 7, clock_flip
8540 Rotate by 90 degrees clockwise and vertically flip, that is:
8548 For values between 4-7, the transposition is only done if the input
8549 video geometry is portrait and not landscape. These values are
8550 deprecated, the @code{passthrough} option should be used instead.
8552 Numerical values are deprecated, and should be dropped in favor of
8556 Do not apply the transposition if the input geometry matches the one
8557 specified by the specified value. It accepts the following values:
8560 Always apply transposition.
8562 Preserve portrait geometry (when @var{height} >= @var{width}).
8564 Preserve landscape geometry (when @var{width} >= @var{height}).
8567 Default value is @code{none}.
8570 For example to rotate by 90 degrees clockwise and preserve portrait
8573 transpose=dir=1:passthrough=portrait
8576 The command above can also be specified as:
8578 transpose=1:portrait
8582 Trim the input so that the output contains one continuous subpart of the input.
8584 It accepts the following parameters:
8587 Specify the time of the start of the kept section, i.e. the frame with the
8588 timestamp @var{start} will be the first frame in the output.
8591 Specify the time of the first frame that will be dropped, i.e. the frame
8592 immediately preceding the one with the timestamp @var{end} will be the last
8593 frame in the output.
8596 This is the same as @var{start}, except this option sets the start timestamp
8597 in timebase units instead of seconds.
8600 This is the same as @var{end}, except this option sets the end timestamp
8601 in timebase units instead of seconds.
8604 The maximum duration of the output in seconds.
8607 The number of the first frame that should be passed to the output.
8610 The number of the first frame that should be dropped.
8613 @option{start}, @option{end}, and @option{duration} are expressed as time
8614 duration specifications; see
8615 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
8616 for the accepted syntax.
8618 Note that the first two sets of the start/end options and the @option{duration}
8619 option look at the frame timestamp, while the _frame variants simply count the
8620 frames that pass through the filter. Also note that this filter does not modify
8621 the timestamps. If you wish for the output timestamps to start at zero, insert a
8622 setpts filter after the trim filter.
8624 If multiple start or end options are set, this filter tries to be greedy and
8625 keep all the frames that match at least one of the specified constraints. To keep
8626 only the part that matches all the constraints at once, chain multiple trim
8629 The defaults are such that all the input is kept. So it is possible to set e.g.
8630 just the end values to keep everything before the specified time.
8635 Drop everything except the second minute of input:
8637 ffmpeg -i INPUT -vf trim=60:120
8641 Keep only the first second:
8643 ffmpeg -i INPUT -vf trim=duration=1
8651 Sharpen or blur the input video.
8653 It accepts the following parameters:
8656 @item luma_msize_x, lx
8657 Set the luma matrix horizontal size. It must be an odd integer between
8658 3 and 63. The default value is 5.
8660 @item luma_msize_y, ly
8661 Set the luma matrix vertical size. It must be an odd integer between 3
8662 and 63. The default value is 5.
8664 @item luma_amount, la
8665 Set the luma effect strength. It must be a floating point number, reasonable
8666 values lay between -1.5 and 1.5.
8668 Negative values will blur the input video, while positive values will
8669 sharpen it, a value of zero will disable the effect.
8671 Default value is 1.0.
8673 @item chroma_msize_x, cx
8674 Set the chroma matrix horizontal size. It must be an odd integer
8675 between 3 and 63. The default value is 5.
8677 @item chroma_msize_y, cy
8678 Set the chroma matrix vertical size. It must be an odd integer
8679 between 3 and 63. The default value is 5.
8681 @item chroma_amount, ca
8682 Set the chroma effect strength. It must be a floating point number, reasonable
8683 values lay between -1.5 and 1.5.
8685 Negative values will blur the input video, while positive values will
8686 sharpen it, a value of zero will disable the effect.
8688 Default value is 0.0.
8691 If set to 1, specify using OpenCL capabilities, only available if
8692 FFmpeg was configured with @code{--enable-opencl}. Default value is 0.
8696 All parameters are optional and default to the equivalent of the
8697 string '5:5:1.0:5:5:0.0'.
8699 @subsection Examples
8703 Apply strong luma sharpen effect:
8705 unsharp=luma_msize_x=7:luma_msize_y=7:luma_amount=2.5
8709 Apply a strong blur of both luma and chroma parameters:
8711 unsharp=7:7:-2:7:7:-2
8715 @anchor{vidstabdetect}
8716 @section vidstabdetect
8718 Analyze video stabilization/deshaking. Perform pass 1 of 2, see
8719 @ref{vidstabtransform} for pass 2.
8721 This filter generates a file with relative translation and rotation
8722 transform information about subsequent frames, which is then used by
8723 the @ref{vidstabtransform} filter.
8725 To enable compilation of this filter you need to configure FFmpeg with
8726 @code{--enable-libvidstab}.
8728 This filter accepts the following options:
8732 Set the path to the file used to write the transforms information.
8733 Default value is @file{transforms.trf}.
8736 Set how shaky the video is and how quick the camera is. It accepts an
8737 integer in the range 1-10, a value of 1 means little shakiness, a
8738 value of 10 means strong shakiness. Default value is 5.
8741 Set the accuracy of the detection process. It must be a value in the
8742 range 1-15. A value of 1 means low accuracy, a value of 15 means high
8743 accuracy. Default value is 15.
8746 Set stepsize of the search process. The region around minimum is
8747 scanned with 1 pixel resolution. Default value is 6.
8750 Set minimum contrast. Below this value a local measurement field is
8751 discarded. Must be a floating point value in the range 0-1. Default
8755 Set reference frame number for tripod mode.
8757 If enabled, the motion of the frames is compared to a reference frame
8758 in the filtered stream, identified by the specified number. The idea
8759 is to compensate all movements in a more-or-less static scene and keep
8760 the camera view absolutely still.
8762 If set to 0, it is disabled. The frames are counted starting from 1.
8765 Show fields and transforms in the resulting frames. It accepts an
8766 integer in the range 0-2. Default value is 0, which disables any
8770 @subsection Examples
8780 Analyze strongly shaky movie and put the results in file
8781 @file{mytransforms.trf}:
8783 vidstabdetect=shakiness=10:accuracy=15:result="mytransforms.trf"
8787 Visualize the result of internal transformations in the resulting
8790 vidstabdetect=show=1
8794 Analyze a video with medium shakiness using @command{ffmpeg}:
8796 ffmpeg -i input -vf vidstabdetect=shakiness=5:show=1 dummy.avi
8800 @anchor{vidstabtransform}
8801 @section vidstabtransform
8803 Video stabilization/deshaking: pass 2 of 2,
8804 see @ref{vidstabdetect} for pass 1.
8806 Read a file with transform information for each frame and
8807 apply/compensate them. Together with the @ref{vidstabdetect}
8808 filter this can be used to deshake videos. See also
8809 @url{http://public.hronopik.de/vid.stab}. It is important to also use
8810 the unsharp filter, see below.
8812 To enable compilation of this filter you need to configure FFmpeg with
8813 @code{--enable-libvidstab}.
8819 Set path to the file used to read the transforms. Default value is
8820 @file{transforms.trf}).
8823 Set the number of frames (value*2 + 1) used for lowpass filtering the
8824 camera movements. Default value is 10.
8826 For example a number of 10 means that 21 frames are used (10 in the
8827 past and 10 in the future) to smoothen the motion in the video. A
8828 larger values leads to a smoother video, but limits the acceleration
8829 of the camera (pan/tilt movements). 0 is a special case where a
8830 static camera is simulated.
8833 Set the camera path optimization algorithm.
8835 Accepted values are:
8838 gaussian kernel low-pass filter on camera motion (default)
8840 averaging on transformations
8844 Set maximal number of pixels to translate frames. Default value is -1,
8848 Set maximal angle in radians (degree*PI/180) to rotate frames. Default
8849 value is -1, meaning no limit.
8852 Specify how to deal with borders that may be visible due to movement
8855 Available values are:
8858 keep image information from previous frame (default)
8860 fill the border black
8864 Invert transforms if set to 1. Default value is 0.
8867 Consider transforms as relative to previsou frame if set to 1,
8868 absolute if set to 0. Default value is 0.
8871 Set percentage to zoom. A positive value will result in a zoom-in
8872 effect, a negative value in a zoom-out effect. Default value is 0 (no
8876 Set optimal zooming to avoid borders.
8878 Accepted values are:
8883 optimal static zoom value is determined (only very strong movements
8884 will lead to visible borders) (default)
8886 optimal adaptive zoom value is determined (no borders will be
8887 visible), see @option{zoomspeed}
8890 Note that the value given at zoom is added to the one calculated here.
8893 Set percent to zoom maximally each frame (enabled when
8894 @option{optzoom} is set to 2). Range is from 0 to 5, default value is
8898 Specify type of interpolation.
8900 Available values are:
8905 linear only horizontal
8907 linear in both directions (default)
8909 cubic in both directions (slow)
8913 Enable virtual tripod mode if set to 1, which is equivalent to
8914 @code{relative=0:smoothing=0}. Default value is 0.
8916 Use also @code{tripod} option of @ref{vidstabdetect}.
8919 Increase log verbosity if set to 1. Also the detected global motions
8920 are written to the temporary file @file{global_motions.trf}. Default
8924 @subsection Examples
8928 Use @command{ffmpeg} for a typical stabilization with default values:
8930 ffmpeg -i inp.mpeg -vf vidstabtransform,unsharp=5:5:0.8:3:3:0.4 inp_stabilized.mpeg
8933 Note the use of the unsharp filter which is always recommended.
8936 Zoom in a bit more and load transform data from a given file:
8938 vidstabtransform=zoom=5:input="mytransforms.trf"
8942 Smoothen the video even more:
8944 vidstabtransform=smoothing=30
8950 Flip the input video vertically.
8952 For example, to vertically flip a video with @command{ffmpeg}:
8954 ffmpeg -i in.avi -vf "vflip" out.avi
8960 Make or reverse a natural vignetting effect.
8962 The filter accepts the following options:
8966 Set lens angle expression as a number of radians.
8968 The value is clipped in the @code{[0,PI/2]} range.
8970 Default value: @code{"PI/5"}
8974 Set center coordinates expressions. Respectively @code{"w/2"} and @code{"h/2"}
8978 Set forward/backward mode.
8980 Available modes are:
8983 The larger the distance from the central point, the darker the image becomes.
8986 The larger the distance from the central point, the brighter the image becomes.
8987 This can be used to reverse a vignette effect, though there is no automatic
8988 detection to extract the lens @option{angle} and other settings (yet). It can
8989 also be used to create a burning effect.
8992 Default value is @samp{forward}.
8995 Set evaluation mode for the expressions (@option{angle}, @option{x0}, @option{y0}).
8997 It accepts the following values:
9000 Evaluate expressions only once during the filter initialization.
9003 Evaluate expressions for each incoming frame. This is way slower than the
9004 @samp{init} mode since it requires all the scalers to be re-computed, but it
9005 allows advanced dynamic expressions.
9008 Default value is @samp{init}.
9011 Set dithering to reduce the circular banding effects. Default is @code{1}
9015 Set vignette aspect. This setting allows one to adjust the shape of the vignette.
9016 Setting this value to the SAR of the input will make a rectangular vignetting
9017 following the dimensions of the video.
9019 Default is @code{1/1}.
9022 @subsection Expressions
9024 The @option{alpha}, @option{x0} and @option{y0} expressions can contain the
9025 following parameters.
9030 input width and height
9033 the number of input frame, starting from 0
9036 the PTS (Presentation TimeStamp) time of the filtered video frame, expressed in
9037 @var{TB} units, NAN if undefined
9040 frame rate of the input video, NAN if the input frame rate is unknown
9043 the PTS (Presentation TimeStamp) of the filtered video frame,
9044 expressed in seconds, NAN if undefined
9047 time base of the input video
9051 @subsection Examples
9055 Apply simple strong vignetting effect:
9061 Make a flickering vignetting:
9063 vignette='PI/4+random(1)*PI/50':eval=frame
9070 Deinterlace the input video ("w3fdif" stands for "Weston 3 Field
9071 Deinterlacing Filter").
9073 Based on the process described by Martin Weston for BBC R&D, and
9074 implemented based on the de-interlace algorithm written by Jim
9075 Easterbrook for BBC R&D, the Weston 3 field deinterlacing filter
9076 uses filter coefficients calculated by BBC R&D.
9078 There are two sets of filter coefficients, so called "simple":
9079 and "complex". Which set of filter coefficients is used can
9080 be set by passing an optional parameter:
9084 Set the interlacing filter coefficients. Accepts one of the following values:
9088 Simple filter coefficient set.
9090 More-complex filter coefficient set.
9092 Default value is @samp{complex}.
9095 Specify which frames to deinterlace. Accept one of the following values:
9099 Deinterlace all frames,
9101 Only deinterlace frames marked as interlaced.
9104 Default value is @samp{all}.
9110 Deinterlace the input video ("yadif" means "yet another deinterlacing
9113 It accepts the following parameters:
9119 The interlacing mode to adopt. It accepts one of the following values:
9123 Output one frame for each frame.
9125 Output one frame for each field.
9126 @item 2, send_frame_nospatial
9127 Like @code{send_frame}, but it skips the spatial interlacing check.
9128 @item 3, send_field_nospatial
9129 Like @code{send_field}, but it skips the spatial interlacing check.
9132 The default value is @code{send_frame}.
9135 The picture field parity assumed for the input interlaced video. It accepts one
9136 of the following values:
9140 Assume the top field is first.
9142 Assume the bottom field is first.
9144 Enable automatic detection of field parity.
9147 The default value is @code{auto}.
9148 If the interlacing is unknown or the decoder does not export this information,
9149 top field first will be assumed.
9152 Specify which frames to deinterlace. Accept one of the following
9157 Deinterlace all frames.
9159 Only deinterlace frames marked as interlaced.
9162 The default value is @code{all}.
9167 Apply Zoom & Pan effect.
9169 This filter accepts the following options:
9173 Set the zoom expression. Default is 1.
9177 Set the x and y expression. Default is 0.
9180 Set the duration expression in number of frames.
9181 This sets for how many number of frames effect will last for
9185 Set the output image size, default is 'hd720'.
9188 Each expression can contain the following constants:
9211 Last calculated 'x' and 'y' position from 'x' and 'y' expression
9212 for current input frame.
9216 'x' and 'y' of last output frame of previous input frame or 0 when there was
9217 not yet such frame (first input frame).
9220 Last calculated zoom from 'z' expression for current input frame.
9223 Last calculated zoom of last output frame of previous input frame.
9226 Number of output frames for current input frame. Calculated from 'd' expression
9227 for each input frame.
9230 number of output frames created for previous input frame
9233 Rational number: input width / input height
9239 display aspect ratio
9243 @subsection Examples
9247 Zoom-in up to 1.5 and pan at same time to some spot near center of picture:
9249 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
9253 @c man end VIDEO FILTERS
9255 @chapter Video Sources
9256 @c man begin VIDEO SOURCES
9258 Below is a description of the currently available video sources.
9262 Buffer video frames, and make them available to the filter chain.
9264 This source is mainly intended for a programmatic use, in particular
9265 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
9267 It accepts the following parameters:
9272 Specify the size (width and height) of the buffered video frames. For the
9273 syntax of this option, check the "Video size" section in the ffmpeg-utils
9277 The input video width.
9280 The input video height.
9283 A string representing the pixel format of the buffered video frames.
9284 It may be a number corresponding to a pixel format, or a pixel format
9288 Specify the timebase assumed by the timestamps of the buffered frames.
9291 Specify the frame rate expected for the video stream.
9293 @item pixel_aspect, sar
9294 The sample (pixel) aspect ratio of the input video.
9297 Specify the optional parameters to be used for the scale filter which
9298 is automatically inserted when an input change is detected in the
9299 input size or format.
9304 buffer=width=320:height=240:pix_fmt=yuv410p:time_base=1/24:sar=1
9307 will instruct the source to accept video frames with size 320x240 and
9308 with format "yuv410p", assuming 1/24 as the timestamps timebase and
9309 square pixels (1:1 sample aspect ratio).
9310 Since the pixel format with name "yuv410p" corresponds to the number 6
9311 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
9312 this example corresponds to:
9314 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
9317 Alternatively, the options can be specified as a flat string, but this
9318 syntax is deprecated:
9320 @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}]
9324 Create a pattern generated by an elementary cellular automaton.
9326 The initial state of the cellular automaton can be defined through the
9327 @option{filename}, and @option{pattern} options. If such options are
9328 not specified an initial state is created randomly.
9330 At each new frame a new row in the video is filled with the result of
9331 the cellular automaton next generation. The behavior when the whole
9332 frame is filled is defined by the @option{scroll} option.
9334 This source accepts the following options:
9338 Read the initial cellular automaton state, i.e. the starting row, from
9340 In the file, each non-whitespace character is considered an alive
9341 cell, a newline will terminate the row, and further characters in the
9342 file will be ignored.
9345 Read the initial cellular automaton state, i.e. the starting row, from
9346 the specified string.
9348 Each non-whitespace character in the string is considered an alive
9349 cell, a newline will terminate the row, and further characters in the
9350 string will be ignored.
9353 Set the video rate, that is the number of frames generated per second.
9356 @item random_fill_ratio, ratio
9357 Set the random fill ratio for the initial cellular automaton row. It
9358 is a floating point number value ranging from 0 to 1, defaults to
9361 This option is ignored when a file or a pattern is specified.
9363 @item random_seed, seed
9364 Set the seed for filling randomly the initial row, must be an integer
9365 included between 0 and UINT32_MAX. If not specified, or if explicitly
9366 set to -1, the filter will try to use a good random seed on a best
9370 Set the cellular automaton rule, it is a number ranging from 0 to 255.
9371 Default value is 110.
9374 Set the size of the output video. For the syntax of this option, check
9375 the "Video size" section in the ffmpeg-utils manual.
9377 If @option{filename} or @option{pattern} is specified, the size is set
9378 by default to the width of the specified initial state row, and the
9379 height is set to @var{width} * PHI.
9381 If @option{size} is set, it must contain the width of the specified
9382 pattern string, and the specified pattern will be centered in the
9385 If a filename or a pattern string is not specified, the size value
9386 defaults to "320x518" (used for a randomly generated initial state).
9389 If set to 1, scroll the output upward when all the rows in the output
9390 have been already filled. If set to 0, the new generated row will be
9391 written over the top row just after the bottom row is filled.
9394 @item start_full, full
9395 If set to 1, completely fill the output with generated rows before
9396 outputting the first frame.
9397 This is the default behavior, for disabling set the value to 0.
9400 If set to 1, stitch the left and right row edges together.
9401 This is the default behavior, for disabling set the value to 0.
9404 @subsection Examples
9408 Read the initial state from @file{pattern}, and specify an output of
9411 cellauto=f=pattern:s=200x400
9415 Generate a random initial row with a width of 200 cells, with a fill
9418 cellauto=ratio=2/3:s=200x200
9422 Create a pattern generated by rule 18 starting by a single alive cell
9423 centered on an initial row with width 100:
9425 cellauto=p=@@:s=100x400:full=0:rule=18
9429 Specify a more elaborated initial pattern:
9431 cellauto=p='@@@@ @@ @@@@':s=100x400:full=0:rule=18
9438 Generate a Mandelbrot set fractal, and progressively zoom towards the
9439 point specified with @var{start_x} and @var{start_y}.
9441 This source accepts the following options:
9446 Set the terminal pts value. Default value is 400.
9449 Set the terminal scale value.
9450 Must be a floating point value. Default value is 0.3.
9453 Set the inner coloring mode, that is the algorithm used to draw the
9454 Mandelbrot fractal internal region.
9456 It shall assume one of the following values:
9461 Show time until convergence.
9463 Set color based on point closest to the origin of the iterations.
9468 Default value is @var{mincol}.
9471 Set the bailout value. Default value is 10.0.
9474 Set the maximum of iterations performed by the rendering
9475 algorithm. Default value is 7189.
9478 Set outer coloring mode.
9479 It shall assume one of following values:
9481 @item iteration_count
9482 Set iteration cound mode.
9483 @item normalized_iteration_count
9484 set normalized iteration count mode.
9486 Default value is @var{normalized_iteration_count}.
9489 Set frame rate, expressed as number of frames per second. Default
9493 Set frame size. For the syntax of this option, check the "Video
9494 size" section in the ffmpeg-utils manual. Default value is "640x480".
9497 Set the initial scale value. Default value is 3.0.
9500 Set the initial x position. Must be a floating point value between
9501 -100 and 100. Default value is -0.743643887037158704752191506114774.
9504 Set the initial y position. Must be a floating point value between
9505 -100 and 100. Default value is -0.131825904205311970493132056385139.
9510 Generate various test patterns, as generated by the MPlayer test filter.
9512 The size of the generated video is fixed, and is 256x256.
9513 This source is useful in particular for testing encoding features.
9515 This source accepts the following options:
9520 Specify the frame rate of the sourced video, as the number of frames
9521 generated per second. It has to be a string in the format
9522 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
9523 number or a valid video frame rate abbreviation. The default value is
9527 Set the duration of the sourced video. See
9528 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
9529 for the accepted syntax.
9531 If not specified, or the expressed duration is negative, the video is
9532 supposed to be generated forever.
9536 Set the number or the name of the test to perform. Supported tests are:
9552 Default value is "all", which will cycle through the list of all tests.
9560 will generate a "dc_luma" test pattern.
9564 Provide a frei0r source.
9566 To enable compilation of this filter you need to install the frei0r
9567 header and configure FFmpeg with @code{--enable-frei0r}.
9569 This source accepts the following parameters:
9574 The size of the video to generate. For the syntax of this option, check the
9575 "Video size" section in the ffmpeg-utils manual.
9578 The framerate of the generated video. It may be a string of the form
9579 @var{num}/@var{den} or a frame rate abbreviation.
9582 The name to the frei0r source to load. For more information regarding frei0r and
9583 how to set the parameters, read the @ref{frei0r} section in the video filters
9587 A '|'-separated list of parameters to pass to the frei0r source.
9591 For example, to generate a frei0r partik0l source with size 200x200
9592 and frame rate 10 which is overlayed on the overlay filter main input:
9594 frei0r_src=size=200x200:framerate=10:filter_name=partik0l:filter_params=1234 [overlay]; [in][overlay] overlay
9599 Generate a life pattern.
9601 This source is based on a generalization of John Conway's life game.
9603 The sourced input represents a life grid, each pixel represents a cell
9604 which can be in one of two possible states, alive or dead. Every cell
9605 interacts with its eight neighbours, which are the cells that are
9606 horizontally, vertically, or diagonally adjacent.
9608 At each interaction the grid evolves according to the adopted rule,
9609 which specifies the number of neighbor alive cells which will make a
9610 cell stay alive or born. The @option{rule} option allows one to specify
9613 This source accepts the following options:
9617 Set the file from which to read the initial grid state. In the file,
9618 each non-whitespace character is considered an alive cell, and newline
9619 is used to delimit the end of each row.
9621 If this option is not specified, the initial grid is generated
9625 Set the video rate, that is the number of frames generated per second.
9628 @item random_fill_ratio, ratio
9629 Set the random fill ratio for the initial random grid. It is a
9630 floating point number value ranging from 0 to 1, defaults to 1/PHI.
9631 It is ignored when a file is specified.
9633 @item random_seed, seed
9634 Set the seed for filling the initial random grid, must be an integer
9635 included between 0 and UINT32_MAX. If not specified, or if explicitly
9636 set to -1, the filter will try to use a good random seed on a best
9642 A rule can be specified with a code of the kind "S@var{NS}/B@var{NB}",
9643 where @var{NS} and @var{NB} are sequences of numbers in the range 0-8,
9644 @var{NS} specifies the number of alive neighbor cells which make a
9645 live cell stay alive, and @var{NB} the number of alive neighbor cells
9646 which make a dead cell to become alive (i.e. to "born").
9647 "s" and "b" can be used in place of "S" and "B", respectively.
9649 Alternatively a rule can be specified by an 18-bits integer. The 9
9650 high order bits are used to encode the next cell state if it is alive
9651 for each number of neighbor alive cells, the low order bits specify
9652 the rule for "borning" new cells. Higher order bits encode for an
9653 higher number of neighbor cells.
9654 For example the number 6153 = @code{(12<<9)+9} specifies a stay alive
9655 rule of 12 and a born rule of 9, which corresponds to "S23/B03".
9657 Default value is "S23/B3", which is the original Conway's game of life
9658 rule, and will keep a cell alive if it has 2 or 3 neighbor alive
9659 cells, and will born a new cell if there are three alive cells around
9663 Set the size of the output video. For the syntax of this option, check the
9664 "Video size" section in the ffmpeg-utils manual.
9666 If @option{filename} is specified, the size is set by default to the
9667 same size of the input file. If @option{size} is set, it must contain
9668 the size specified in the input file, and the initial grid defined in
9669 that file is centered in the larger resulting area.
9671 If a filename is not specified, the size value defaults to "320x240"
9672 (used for a randomly generated initial grid).
9675 If set to 1, stitch the left and right grid edges together, and the
9676 top and bottom edges also. Defaults to 1.
9679 Set cell mold speed. If set, a dead cell will go from @option{death_color} to
9680 @option{mold_color} with a step of @option{mold}. @option{mold} can have a
9681 value from 0 to 255.
9684 Set the color of living (or new born) cells.
9687 Set the color of dead cells. If @option{mold} is set, this is the first color
9688 used to represent a dead cell.
9691 Set mold color, for definitely dead and moldy cells.
9693 For the syntax of these 3 color options, check the "Color" section in the
9694 ffmpeg-utils manual.
9697 @subsection Examples
9701 Read a grid from @file{pattern}, and center it on a grid of size
9704 life=f=pattern:s=300x300
9708 Generate a random grid of size 200x200, with a fill ratio of 2/3:
9710 life=ratio=2/3:s=200x200
9714 Specify a custom rule for evolving a randomly generated grid:
9720 Full example with slow death effect (mold) using @command{ffplay}:
9722 ffplay -f lavfi life=s=300x200:mold=10:r=60:ratio=0.1:death_color=#C83232:life_color=#00ff00,scale=1200:800:flags=16
9727 @anchor{haldclutsrc}
9731 @anchor{smptehdbars}
9733 @section color, haldclutsrc, nullsrc, rgbtestsrc, smptebars, smptehdbars, testsrc
9735 The @code{color} source provides an uniformly colored input.
9737 The @code{haldclutsrc} source provides an identity Hald CLUT. See also
9738 @ref{haldclut} filter.
9740 The @code{nullsrc} source returns unprocessed video frames. It is
9741 mainly useful to be employed in analysis / debugging tools, or as the
9742 source for filters which ignore the input data.
9744 The @code{rgbtestsrc} source generates an RGB test pattern useful for
9745 detecting RGB vs BGR issues. You should see a red, green and blue
9746 stripe from top to bottom.
9748 The @code{smptebars} source generates a color bars pattern, based on
9749 the SMPTE Engineering Guideline EG 1-1990.
9751 The @code{smptehdbars} source generates a color bars pattern, based on
9752 the SMPTE RP 219-2002.
9754 The @code{testsrc} source generates a test video pattern, showing a
9755 color pattern, a scrolling gradient and a timestamp. This is mainly
9756 intended for testing purposes.
9758 The sources accept the following parameters:
9763 Specify the color of the source, only available in the @code{color}
9764 source. For the syntax of this option, check the "Color" section in the
9765 ffmpeg-utils manual.
9768 Specify the level of the Hald CLUT, only available in the @code{haldclutsrc}
9769 source. A level of @code{N} generates a picture of @code{N*N*N} by @code{N*N*N}
9770 pixels to be used as identity matrix for 3D lookup tables. Each component is
9771 coded on a @code{1/(N*N)} scale.
9774 Specify the size of the sourced video. For the syntax of this option, check the
9775 "Video size" section in the ffmpeg-utils manual. The default value is
9778 This option is not available with the @code{haldclutsrc} filter.
9781 Specify the frame rate of the sourced video, as the number of frames
9782 generated per second. It has to be a string in the format
9783 @var{frame_rate_num}/@var{frame_rate_den}, an integer number, a floating point
9784 number or a valid video frame rate abbreviation. The default value is
9788 Set the sample aspect ratio of the sourced video.
9791 Set the duration of the sourced video. See
9792 @ref{time duration syntax,,the Time duration section in the ffmpeg-utils(1) manual,ffmpeg-utils}
9793 for the accepted syntax.
9795 If not specified, or the expressed duration is negative, the video is
9796 supposed to be generated forever.
9799 Set the number of decimals to show in the timestamp, only available in the
9800 @code{testsrc} source.
9802 The displayed timestamp value will correspond to the original
9803 timestamp value multiplied by the power of 10 of the specified
9804 value. Default value is 0.
9807 For example the following:
9809 testsrc=duration=5.3:size=qcif:rate=10
9812 will generate a video with a duration of 5.3 seconds, with size
9813 176x144 and a frame rate of 10 frames per second.
9815 The following graph description will generate a red source
9816 with an opacity of 0.2, with size "qcif" and a frame rate of 10
9819 color=c=red@@0.2:s=qcif:r=10
9822 If the input content is to be ignored, @code{nullsrc} can be used. The
9823 following command generates noise in the luminance plane by employing
9824 the @code{geq} filter:
9826 nullsrc=s=256x256, geq=random(1)*255:128:128
9829 @subsection Commands
9831 The @code{color} source supports the following commands:
9835 Set the color of the created image. Accepts the same syntax of the
9836 corresponding @option{color} option.
9839 @c man end VIDEO SOURCES
9841 @chapter Video Sinks
9842 @c man begin VIDEO SINKS
9844 Below is a description of the currently available video sinks.
9848 Buffer video frames, and make them available to the end of the filter
9851 This sink is mainly intended for programmatic use, in particular
9852 through the interface defined in @file{libavfilter/buffersink.h}
9853 or the options system.
9855 It accepts a pointer to an AVBufferSinkContext structure, which
9856 defines the incoming buffers' formats, to be passed as the opaque
9857 parameter to @code{avfilter_init_filter} for initialization.
9861 Null video sink: do absolutely nothing with the input video. It is
9862 mainly useful as a template and for use in analysis / debugging
9865 @c man end VIDEO SINKS
9867 @chapter Multimedia Filters
9868 @c man begin MULTIMEDIA FILTERS
9870 Below is a description of the currently available multimedia filters.
9872 @section avectorscope
9874 Convert input audio to a video output, representing the audio vector
9877 The filter is used to measure the difference between channels of stereo
9878 audio stream. A monoaural signal, consisting of identical left and right
9879 signal, results in straight vertical line. Any stereo separation is visible
9880 as a deviation from this line, creating a Lissajous figure.
9881 If the straight (or deviation from it) but horizontal line appears this
9882 indicates that the left and right channels are out of phase.
9884 The filter accepts the following options:
9888 Set the vectorscope mode.
9890 Available values are:
9893 Lissajous rotated by 45 degrees.
9896 Same as above but not rotated.
9899 Default value is @samp{lissajous}.
9902 Set the video size for the output. For the syntax of this option, check the "Video size"
9903 section in the ffmpeg-utils manual. Default value is @code{400x400}.
9906 Set the output frame rate. Default value is @code{25}.
9911 Specify the red, green and blue contrast. Default values are @code{40}, @code{160} and @code{80}.
9912 Allowed range is @code{[0, 255]}.
9917 Specify the red, green and blue fade. Default values are @code{15}, @code{10} and @code{5}.
9918 Allowed range is @code{[0, 255]}.
9921 Set the zoom factor. Default value is @code{1}. Allowed range is @code{[1, 10]}.
9924 @subsection Examples
9928 Complete example using @command{ffplay}:
9930 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
9931 [a] avectorscope=zoom=1.3:rc=2:gc=200:bc=10:rf=1:gf=8:bf=7 [out0]'
9937 Concatenate audio and video streams, joining them together one after the
9940 The filter works on segments of synchronized video and audio streams. All
9941 segments must have the same number of streams of each type, and that will
9942 also be the number of streams at output.
9944 The filter accepts the following options:
9949 Set the number of segments. Default is 2.
9952 Set the number of output video streams, that is also the number of video
9953 streams in each segment. Default is 1.
9956 Set the number of output audio streams, that is also the number of audio
9957 streams in each segment. Default is 0.
9960 Activate unsafe mode: do not fail if segments have a different format.
9964 The filter has @var{v}+@var{a} outputs: first @var{v} video outputs, then
9965 @var{a} audio outputs.
9967 There are @var{n}x(@var{v}+@var{a}) inputs: first the inputs for the first
9968 segment, in the same order as the outputs, then the inputs for the second
9971 Related streams do not always have exactly the same duration, for various
9972 reasons including codec frame size or sloppy authoring. For that reason,
9973 related synchronized streams (e.g. a video and its audio track) should be
9974 concatenated at once. The concat filter will use the duration of the longest
9975 stream in each segment (except the last one), and if necessary pad shorter
9976 audio streams with silence.
9978 For this filter to work correctly, all segments must start at timestamp 0.
9980 All corresponding streams must have the same parameters in all segments; the
9981 filtering system will automatically select a common pixel format for video
9982 streams, and a common sample format, sample rate and channel layout for
9983 audio streams, but other settings, such as resolution, must be converted
9984 explicitly by the user.
9986 Different frame rates are acceptable but will result in variable frame rate
9987 at output; be sure to configure the output file to handle it.
9989 @subsection Examples
9993 Concatenate an opening, an episode and an ending, all in bilingual version
9994 (video in stream 0, audio in streams 1 and 2):
9996 ffmpeg -i opening.mkv -i episode.mkv -i ending.mkv -filter_complex \
9997 '[0:0] [0:1] [0:2] [1:0] [1:1] [1:2] [2:0] [2:1] [2:2]
9998 concat=n=3:v=1:a=2 [v] [a1] [a2]' \
9999 -map '[v]' -map '[a1]' -map '[a2]' output.mkv
10003 Concatenate two parts, handling audio and video separately, using the
10004 (a)movie sources, and adjusting the resolution:
10006 movie=part1.mp4, scale=512:288 [v1] ; amovie=part1.mp4 [a1] ;
10007 movie=part2.mp4, scale=512:288 [v2] ; amovie=part2.mp4 [a2] ;
10008 [v1] [v2] concat [outv] ; [a1] [a2] concat=v=0:a=1 [outa]
10010 Note that a desync will happen at the stitch if the audio and video streams
10011 do not have exactly the same duration in the first file.
10017 EBU R128 scanner filter. This filter takes an audio stream as input and outputs
10018 it unchanged. By default, it logs a message at a frequency of 10Hz with the
10019 Momentary loudness (identified by @code{M}), Short-term loudness (@code{S}),
10020 Integrated loudness (@code{I}) and Loudness Range (@code{LRA}).
10022 The filter also has a video output (see the @var{video} option) with a real
10023 time graph to observe the loudness evolution. The graphic contains the logged
10024 message mentioned above, so it is not printed anymore when this option is set,
10025 unless the verbose logging is set. The main graphing area contains the
10026 short-term loudness (3 seconds of analysis), and the gauge on the right is for
10027 the momentary loudness (400 milliseconds).
10029 More information about the Loudness Recommendation EBU R128 on
10030 @url{http://tech.ebu.ch/loudness}.
10032 The filter accepts the following options:
10037 Activate the video output. The audio stream is passed unchanged whether this
10038 option is set or no. The video stream will be the first output stream if
10039 activated. Default is @code{0}.
10042 Set the video size. This option is for video only. For the syntax of this
10043 option, check the "Video size" section in the ffmpeg-utils manual. Default
10044 and minimum resolution is @code{640x480}.
10047 Set the EBU scale meter. Default is @code{9}. Common values are @code{9} and
10048 @code{18}, respectively for EBU scale meter +9 and EBU scale meter +18. Any
10049 other integer value between this range is allowed.
10052 Set metadata injection. If set to @code{1}, the audio input will be segmented
10053 into 100ms output frames, each of them containing various loudness information
10054 in metadata. All the metadata keys are prefixed with @code{lavfi.r128.}.
10056 Default is @code{0}.
10059 Force the frame logging level.
10061 Available values are:
10064 information logging level
10066 verbose logging level
10069 By default, the logging level is set to @var{info}. If the @option{video} or
10070 the @option{metadata} options are set, it switches to @var{verbose}.
10075 Available modes can be cumulated (the option is a @code{flag} type). Possible
10079 Disable any peak mode (default).
10081 Enable sample-peak mode.
10083 Simple peak mode looking for the higher sample value. It logs a message
10084 for sample-peak (identified by @code{SPK}).
10086 Enable true-peak mode.
10088 If enabled, the peak lookup is done on an over-sampled version of the input
10089 stream for better peak accuracy. It logs a message for true-peak.
10090 (identified by @code{TPK}) and true-peak per frame (identified by @code{FTPK}).
10091 This mode requires a build with @code{libswresample}.
10096 @subsection Examples
10100 Real-time graph using @command{ffplay}, with a EBU scale meter +18:
10102 ffplay -f lavfi -i "amovie=input.mp3,ebur128=video=1:meter=18 [out0][out1]"
10106 Run an analysis with @command{ffmpeg}:
10108 ffmpeg -nostats -i input.mp3 -filter_complex ebur128 -f null -
10112 @section interleave, ainterleave
10114 Temporally interleave frames from several inputs.
10116 @code{interleave} works with video inputs, @code{ainterleave} with audio.
10118 These filters read frames from several inputs and send the oldest
10119 queued frame to the output.
10121 Input streams must have a well defined, monotonically increasing frame
10124 In order to submit one frame to output, these filters need to enqueue
10125 at least one frame for each input, so they cannot work in case one
10126 input is not yet terminated and will not receive incoming frames.
10128 For example consider the case when one input is a @code{select} filter
10129 which always drop input frames. The @code{interleave} filter will keep
10130 reading from that input, but it will never be able to send new frames
10131 to output until the input will send an end-of-stream signal.
10133 Also, depending on inputs synchronization, the filters will drop
10134 frames in case one input receives more frames than the other ones, and
10135 the queue is already filled.
10137 These filters accept the following options:
10141 Set the number of different inputs, it is 2 by default.
10144 @subsection Examples
10148 Interleave frames belonging to different streams using @command{ffmpeg}:
10150 ffmpeg -i bambi.avi -i pr0n.mkv -filter_complex "[0:v][1:v] interleave" out.avi
10154 Add flickering blur effect:
10156 select='if(gt(random(0), 0.2), 1, 2)':n=2 [tmp], boxblur=2:2, [tmp] interleave
10160 @section perms, aperms
10162 Set read/write permissions for the output frames.
10164 These filters are mainly aimed at developers to test direct path in the
10165 following filter in the filtergraph.
10167 The filters accept the following options:
10171 Select the permissions mode.
10173 It accepts the following values:
10176 Do nothing. This is the default.
10178 Set all the output frames read-only.
10180 Set all the output frames directly writable.
10182 Make the frame read-only if writable, and writable if read-only.
10184 Set each output frame read-only or writable randomly.
10188 Set the seed for the @var{random} mode, must be an integer included between
10189 @code{0} and @code{UINT32_MAX}. If not specified, or if explicitly set to
10190 @code{-1}, the filter will try to use a good random seed on a best effort
10194 Note: in case of auto-inserted filter between the permission filter and the
10195 following one, the permission might not be received as expected in that
10196 following filter. Inserting a @ref{format} or @ref{aformat} filter before the
10197 perms/aperms filter can avoid this problem.
10199 @section select, aselect
10201 Select frames to pass in output.
10203 This filter accepts the following options:
10208 Set expression, which is evaluated for each input frame.
10210 If the expression is evaluated to zero, the frame is discarded.
10212 If the evaluation result is negative or NaN, the frame is sent to the
10213 first output; otherwise it is sent to the output with index
10214 @code{ceil(val)-1}, assuming that the input index starts from 0.
10216 For example a value of @code{1.2} corresponds to the output with index
10217 @code{ceil(1.2)-1 = 2-1 = 1}, that is the second output.
10220 Set the number of outputs. The output to which to send the selected
10221 frame is based on the result of the evaluation. Default value is 1.
10224 The expression can contain the following constants:
10228 The (sequential) number of the filtered frame, starting from 0.
10231 The (sequential) number of the selected frame, starting from 0.
10233 @item prev_selected_n
10234 The sequential number of the last selected frame. It's NAN if undefined.
10237 The timebase of the input timestamps.
10240 The PTS (Presentation TimeStamp) of the filtered video frame,
10241 expressed in @var{TB} units. It's NAN if undefined.
10244 The PTS of the filtered video frame,
10245 expressed in seconds. It's NAN if undefined.
10248 The PTS of the previously filtered video frame. It's NAN if undefined.
10250 @item prev_selected_pts
10251 The PTS of the last previously filtered video frame. It's NAN if undefined.
10253 @item prev_selected_t
10254 The PTS of the last previously selected video frame. It's NAN if undefined.
10257 The PTS of the first video frame in the video. It's NAN if undefined.
10260 The time of the first video frame in the video. It's NAN if undefined.
10262 @item pict_type @emph{(video only)}
10263 The type of the filtered frame. It can assume one of the following
10275 @item interlace_type @emph{(video only)}
10276 The frame interlace type. It can assume one of the following values:
10279 The frame is progressive (not interlaced).
10281 The frame is top-field-first.
10283 The frame is bottom-field-first.
10286 @item consumed_sample_n @emph{(audio only)}
10287 the number of selected samples before the current frame
10289 @item samples_n @emph{(audio only)}
10290 the number of samples in the current frame
10292 @item sample_rate @emph{(audio only)}
10293 the input sample rate
10296 This is 1 if the filtered frame is a key-frame, 0 otherwise.
10299 the position in the file of the filtered frame, -1 if the information
10300 is not available (e.g. for synthetic video)
10302 @item scene @emph{(video only)}
10303 value between 0 and 1 to indicate a new scene; a low value reflects a low
10304 probability for the current frame to introduce a new scene, while a higher
10305 value means the current frame is more likely to be one (see the example below)
10309 The default value of the select expression is "1".
10311 @subsection Examples
10315 Select all frames in input:
10320 The example above is the same as:
10332 Select only I-frames:
10334 select='eq(pict_type\,I)'
10338 Select one frame every 100:
10340 select='not(mod(n\,100))'
10344 Select only frames contained in the 10-20 time interval:
10346 select=between(t\,10\,20)
10350 Select only I frames contained in the 10-20 time interval:
10352 select=between(t\,10\,20)*eq(pict_type\,I)
10356 Select frames with a minimum distance of 10 seconds:
10358 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
10362 Use aselect to select only audio frames with samples number > 100:
10364 aselect='gt(samples_n\,100)'
10368 Create a mosaic of the first scenes:
10370 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
10373 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
10377 Send even and odd frames to separate outputs, and compose them:
10379 select=n=2:e='mod(n, 2)+1' [odd][even]; [odd] pad=h=2*ih [tmp]; [tmp][even] overlay=y=h
10383 @section sendcmd, asendcmd
10385 Send commands to filters in the filtergraph.
10387 These filters read commands to be sent to other filters in the
10390 @code{sendcmd} must be inserted between two video filters,
10391 @code{asendcmd} must be inserted between two audio filters, but apart
10392 from that they act the same way.
10394 The specification of commands can be provided in the filter arguments
10395 with the @var{commands} option, or in a file specified by the
10396 @var{filename} option.
10398 These filters accept the following options:
10401 Set the commands to be read and sent to the other filters.
10403 Set the filename of the commands to be read and sent to the other
10407 @subsection Commands syntax
10409 A commands description consists of a sequence of interval
10410 specifications, comprising a list of commands to be executed when a
10411 particular event related to that interval occurs. The occurring event
10412 is typically the current frame time entering or leaving a given time
10415 An interval is specified by the following syntax:
10417 @var{START}[-@var{END}] @var{COMMANDS};
10420 The time interval is specified by the @var{START} and @var{END} times.
10421 @var{END} is optional and defaults to the maximum time.
10423 The current frame time is considered within the specified interval if
10424 it is included in the interval [@var{START}, @var{END}), that is when
10425 the time is greater or equal to @var{START} and is lesser than
10428 @var{COMMANDS} consists of a sequence of one or more command
10429 specifications, separated by ",", relating to that interval. The
10430 syntax of a command specification is given by:
10432 [@var{FLAGS}] @var{TARGET} @var{COMMAND} @var{ARG}
10435 @var{FLAGS} is optional and specifies the type of events relating to
10436 the time interval which enable sending the specified command, and must
10437 be a non-null sequence of identifier flags separated by "+" or "|" and
10438 enclosed between "[" and "]".
10440 The following flags are recognized:
10443 The command is sent when the current frame timestamp enters the
10444 specified interval. In other words, the command is sent when the
10445 previous frame timestamp was not in the given interval, and the
10449 The command is sent when the current frame timestamp leaves the
10450 specified interval. In other words, the command is sent when the
10451 previous frame timestamp was in the given interval, and the
10455 If @var{FLAGS} is not specified, a default value of @code{[enter]} is
10458 @var{TARGET} specifies the target of the command, usually the name of
10459 the filter class or a specific filter instance name.
10461 @var{COMMAND} specifies the name of the command for the target filter.
10463 @var{ARG} is optional and specifies the optional list of argument for
10464 the given @var{COMMAND}.
10466 Between one interval specification and another, whitespaces, or
10467 sequences of characters starting with @code{#} until the end of line,
10468 are ignored and can be used to annotate comments.
10470 A simplified BNF description of the commands specification syntax
10473 @var{COMMAND_FLAG} ::= "enter" | "leave"
10474 @var{COMMAND_FLAGS} ::= @var{COMMAND_FLAG} [(+|"|")@var{COMMAND_FLAG}]
10475 @var{COMMAND} ::= ["[" @var{COMMAND_FLAGS} "]"] @var{TARGET} @var{COMMAND} [@var{ARG}]
10476 @var{COMMANDS} ::= @var{COMMAND} [,@var{COMMANDS}]
10477 @var{INTERVAL} ::= @var{START}[-@var{END}] @var{COMMANDS}
10478 @var{INTERVALS} ::= @var{INTERVAL}[;@var{INTERVALS}]
10481 @subsection Examples
10485 Specify audio tempo change at second 4:
10487 asendcmd=c='4.0 atempo tempo 1.5',atempo
10491 Specify a list of drawtext and hue commands in a file.
10493 # show text in the interval 5-10
10494 5.0-10.0 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=hello world',
10495 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=';
10497 # desaturate the image in the interval 15-20
10498 15.0-20.0 [enter] hue s 0,
10499 [enter] drawtext reinit 'fontfile=FreeSerif.ttf:text=nocolor',
10501 [leave] drawtext reinit 'fontfile=FreeSerif.ttf:text=color';
10503 # apply an exponential saturation fade-out effect, starting from time 25
10504 25 [enter] hue s exp(25-t)
10507 A filtergraph allowing to read and process the above command list
10508 stored in a file @file{test.cmd}, can be specified with:
10510 sendcmd=f=test.cmd,drawtext=fontfile=FreeSerif.ttf:text='',hue
10515 @section setpts, asetpts
10517 Change the PTS (presentation timestamp) of the input frames.
10519 @code{setpts} works on video frames, @code{asetpts} on audio frames.
10521 This filter accepts the following options:
10526 The expression which is evaluated for each frame to construct its timestamp.
10530 The expression is evaluated through the eval API and can contain the following
10535 frame rate, only defined for constant frame-rate video
10538 The presentation timestamp in input
10541 The count of the input frame for video or the number of consumed samples,
10542 not including the current frame for audio, starting from 0.
10544 @item NB_CONSUMED_SAMPLES
10545 The number of consumed samples, not including the current frame (only
10548 @item NB_SAMPLES, S
10549 The number of samples in the current frame (only audio)
10551 @item SAMPLE_RATE, SR
10552 The audio sample rate.
10555 The PTS of the first frame.
10558 the time in seconds of the first frame
10561 State whether the current frame is interlaced.
10564 the time in seconds of the current frame
10567 original position in the file of the frame, or undefined if undefined
10568 for the current frame
10571 The previous input PTS.
10574 previous input time in seconds
10577 The previous output PTS.
10580 previous output time in seconds
10583 The wallclock (RTC) time in microseconds.. This is deprecated, use time(0)
10587 The wallclock (RTC) time at the start of the movie in microseconds.
10590 The timebase of the input timestamps.
10594 @subsection Examples
10598 Start counting PTS from zero
10600 setpts=PTS-STARTPTS
10604 Apply fast motion effect:
10610 Apply slow motion effect:
10616 Set fixed rate of 25 frames per second:
10622 Set fixed rate 25 fps with some jitter:
10624 setpts='1/(25*TB) * (N + 0.05 * sin(N*2*PI/25))'
10628 Apply an offset of 10 seconds to the input PTS:
10634 Generate timestamps from a "live source" and rebase onto the current timebase:
10636 setpts='(RTCTIME - RTCSTART) / (TB * 1000000)'
10640 Generate timestamps by counting samples:
10647 @section settb, asettb
10649 Set the timebase to use for the output frames timestamps.
10650 It is mainly useful for testing timebase configuration.
10652 It accepts the following parameters:
10657 The expression which is evaluated into the output timebase.
10661 The value for @option{tb} is an arithmetic expression representing a
10662 rational. The expression can contain the constants "AVTB" (the default
10663 timebase), "intb" (the input timebase) and "sr" (the sample rate,
10664 audio only). Default value is "intb".
10666 @subsection Examples
10670 Set the timebase to 1/25:
10676 Set the timebase to 1/10:
10682 Set the timebase to 1001/1000:
10688 Set the timebase to 2*intb:
10694 Set the default timebase value:
10701 Convert input audio to a video output representing
10702 frequency spectrum logarithmically (using constant Q transform with
10703 Brown-Puckette algorithm), with musical tone scale, from E0 to D#10 (10 octaves).
10705 The filter accepts the following options:
10709 Specify transform volume (multiplier) expression. The expression can contain
10712 @item frequency, freq, f
10713 the frequency where transform is evaluated
10714 @item timeclamp, tc
10715 value of timeclamp option
10719 @item a_weighting(f)
10720 A-weighting of equal loudness
10721 @item b_weighting(f)
10722 B-weighting of equal loudness
10723 @item c_weighting(f)
10724 C-weighting of equal loudness
10726 Default value is @code{16}.
10729 Specify transform length expression. The expression can contain variables:
10731 @item frequency, freq, f
10732 the frequency where transform is evaluated
10733 @item timeclamp, tc
10734 value of timeclamp option
10736 Default value is @code{384/f*tc/(384/f+tc)}.
10739 Specify the transform timeclamp. At low frequency, there is trade-off between
10740 accuracy in time domain and frequency domain. If timeclamp is lower,
10741 event in time domain is represented more accurately (such as fast bass drum),
10742 otherwise event in frequency domain is represented more accurately
10743 (such as bass guitar). Acceptable value is [0.1, 1.0]. Default value is @code{0.17}.
10746 Specify the transform coeffclamp. If coeffclamp is lower, transform is
10747 more accurate, otherwise transform is faster. Acceptable value is [0.1, 10.0].
10748 Default value is @code{1.0}.
10751 Specify gamma. Lower gamma makes the spectrum more contrast, higher gamma
10752 makes the spectrum having more range. Acceptable value is [1.0, 7.0].
10753 Default value is @code{3.0}.
10756 Specify font file for use with freetype. If not specified, use embedded font.
10759 Specify font color expression. This is arithmetic expression that should return
10760 integer value 0xRRGGBB. The expression can contain variables:
10762 @item frequency, freq, f
10763 the frequency where transform is evaluated
10764 @item timeclamp, tc
10765 value of timeclamp option
10770 midi number of frequency f, some midi numbers: E0(16), C1(24), C2(36), A4(69)
10771 @item r(x), g(x), b(x)
10772 red, green, and blue value of intensity x
10774 Default value is @code{st(0, (midi(f)-59.5)/12);
10775 st(1, if(between(ld(0),0,1), 0.5-0.5*cos(2*PI*ld(0)), 0));
10776 r(1-ld(1)) + b(ld(1))}
10779 If set to 1 (the default), the video size is 1920x1080 (full HD),
10780 if set to 0, the video size is 960x540. Use this option to make CPU usage lower.
10783 Specify video fps. Default value is @code{25}.
10786 Specify number of transform per frame, so there are fps*count transforms
10787 per second. Note that audio data rate must be divisible by fps*count.
10788 Default value is @code{6}.
10792 @subsection Examples
10796 Playing audio while showing the spectrum:
10798 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt [out0]'
10802 Same as above, but with frame rate 30 fps:
10804 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fps=30:count=5 [out0]'
10808 Playing at 960x540 and lower CPU usage:
10810 ffplay -f lavfi 'amovie=a.mp3, asplit [a][out1]; [a] showcqt=fullhd=0:count=3 [out0]'
10814 A1 and its harmonics: A1, A2, (near)E3, A3:
10816 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),
10817 asplit[a][out1]; [a] showcqt [out0]'
10821 Same as above, but with more accuracy in frequency domain (and slower):
10823 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),
10824 asplit[a][out1]; [a] showcqt=timeclamp=0.5 [out0]'
10828 B-weighting of equal loudness
10830 volume=16*b_weighting(f)
10836 tlength=100/f*tc/(100/f+tc)
10840 Custom fontcolor, C-note is colored green, others are colored blue
10842 fontcolor='if(mod(floor(midi(f)+0.5),12), 0x0000FF, g(1))'
10847 @section showspectrum
10849 Convert input audio to a video output, representing the audio frequency
10852 The filter accepts the following options:
10856 Specify the video size for the output. For the syntax of this option, check
10857 the "Video size" section in the ffmpeg-utils manual. Default value is
10861 Specify how the spectrum should slide along the window.
10863 It accepts the following values:
10866 the samples start again on the left when they reach the right
10868 the samples scroll from right to left
10870 frames are only produced when the samples reach the right
10873 Default value is @code{replace}.
10876 Specify display mode.
10878 It accepts the following values:
10881 all channels are displayed in the same row
10883 all channels are displayed in separate rows
10886 Default value is @samp{combined}.
10889 Specify display color mode.
10891 It accepts the following values:
10894 each channel is displayed in a separate color
10896 each channel is is displayed using the same color scheme
10899 Default value is @samp{channel}.
10902 Specify scale used for calculating intensity color values.
10904 It accepts the following values:
10909 square root, default
10916 Default value is @samp{sqrt}.
10919 Set saturation modifier for displayed colors. Negative values provide
10920 alternative color scheme. @code{0} is no saturation at all.
10921 Saturation must be in [-10.0, 10.0] range.
10922 Default value is @code{1}.
10925 Set window function.
10927 It accepts the following values:
10930 No samples pre-processing (do not expect this to be faster)
10939 Default value is @code{hann}.
10942 The usage is very similar to the showwaves filter; see the examples in that
10945 @subsection Examples
10949 Large window with logarithmic color scaling:
10951 showspectrum=s=1280x480:scale=log
10955 Complete example for a colored and sliding spectrum per channel using @command{ffplay}:
10957 ffplay -f lavfi 'amovie=input.mp3, asplit [a][out1];
10958 [a] showspectrum=mode=separate:color=intensity:slide=1:scale=cbrt [out0]'
10964 Convert input audio to a video output, representing the samples waves.
10966 The filter accepts the following options:
10970 Specify the video size for the output. For the syntax of this option, check
10971 the "Video size" section in the ffmpeg-utils manual. Default value
10977 Available values are:
10980 Draw a point for each sample.
10983 Draw a vertical line for each sample.
10986 Draw a point for each sample and a line between them.
10989 Draw a centered vertical line for each sample.
10992 Default value is @code{point}.
10995 Set the number of samples which are printed on the same column. A
10996 larger value will decrease the frame rate. Must be a positive
10997 integer. This option can be set only if the value for @var{rate}
10998 is not explicitly specified.
11001 Set the (approximate) output frame rate. This is done by setting the
11002 option @var{n}. Default value is "25".
11004 @item split_channels
11005 Set if channels should be drawn separately or overlap. Default value is 0.
11009 @subsection Examples
11013 Output the input file audio and the corresponding video representation
11016 amovie=a.mp3,asplit[out0],showwaves[out1]
11020 Create a synthetic signal and show it with showwaves, forcing a
11021 frame rate of 30 frames per second:
11023 aevalsrc=sin(1*2*PI*t)*sin(880*2*PI*t):cos(2*PI*200*t),asplit[out0],showwaves=r=30[out1]
11027 @section split, asplit
11029 Split input into several identical outputs.
11031 @code{asplit} works with audio input, @code{split} with video.
11033 The filter accepts a single parameter which specifies the number of outputs. If
11034 unspecified, it defaults to 2.
11036 @subsection Examples
11040 Create two separate outputs from the same input:
11042 [in] split [out0][out1]
11046 To create 3 or more outputs, you need to specify the number of
11049 [in] asplit=3 [out0][out1][out2]
11053 Create two separate outputs from the same input, one cropped and
11056 [in] split [splitout1][splitout2];
11057 [splitout1] crop=100:100:0:0 [cropout];
11058 [splitout2] pad=200:200:100:100 [padout];
11062 Create 5 copies of the input audio with @command{ffmpeg}:
11064 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
11070 Receive commands sent through a libzmq client, and forward them to
11071 filters in the filtergraph.
11073 @code{zmq} and @code{azmq} work as a pass-through filters. @code{zmq}
11074 must be inserted between two video filters, @code{azmq} between two
11077 To enable these filters you need to install the libzmq library and
11078 headers and configure FFmpeg with @code{--enable-libzmq}.
11080 For more information about libzmq see:
11081 @url{http://www.zeromq.org/}
11083 The @code{zmq} and @code{azmq} filters work as a libzmq server, which
11084 receives messages sent through a network interface defined by the
11085 @option{bind_address} option.
11087 The received message must be in the form:
11089 @var{TARGET} @var{COMMAND} [@var{ARG}]
11092 @var{TARGET} specifies the target of the command, usually the name of
11093 the filter class or a specific filter instance name.
11095 @var{COMMAND} specifies the name of the command for the target filter.
11097 @var{ARG} is optional and specifies the optional argument list for the
11098 given @var{COMMAND}.
11100 Upon reception, the message is processed and the corresponding command
11101 is injected into the filtergraph. Depending on the result, the filter
11102 will send a reply to the client, adopting the format:
11104 @var{ERROR_CODE} @var{ERROR_REASON}
11108 @var{MESSAGE} is optional.
11110 @subsection Examples
11112 Look at @file{tools/zmqsend} for an example of a zmq client which can
11113 be used to send commands processed by these filters.
11115 Consider the following filtergraph generated by @command{ffplay}
11117 ffplay -dumpgraph 1 -f lavfi "
11118 color=s=100x100:c=red [l];
11119 color=s=100x100:c=blue [r];
11120 nullsrc=s=200x100, zmq [bg];
11121 [bg][l] overlay [bg+l];
11122 [bg+l][r] overlay=x=100 "
11125 To change the color of the left side of the video, the following
11126 command can be used:
11128 echo Parsed_color_0 c yellow | tools/zmqsend
11131 To change the right side:
11133 echo Parsed_color_1 c pink | tools/zmqsend
11136 @c man end MULTIMEDIA FILTERS
11138 @chapter Multimedia Sources
11139 @c man begin MULTIMEDIA SOURCES
11141 Below is a description of the currently available multimedia sources.
11145 This is the same as @ref{movie} source, except it selects an audio
11151 Read audio and/or video stream(s) from a movie container.
11153 It accepts the following parameters:
11157 The name of the resource to read (not necessarily a file; it can also be a
11158 device or a stream accessed through some protocol).
11160 @item format_name, f
11161 Specifies the format assumed for the movie to read, and can be either
11162 the name of a container or an input device. If not specified, the
11163 format is guessed from @var{movie_name} or by probing.
11165 @item seek_point, sp
11166 Specifies the seek point in seconds. The frames will be output
11167 starting from this seek point. The parameter is evaluated with
11168 @code{av_strtod}, so the numerical value may be suffixed by an IS
11169 postfix. The default value is "0".
11172 Specifies the streams to read. Several streams can be specified,
11173 separated by "+". The source will then have as many outputs, in the
11174 same order. The syntax is explained in the ``Stream specifiers''
11175 section in the ffmpeg manual. Two special names, "dv" and "da" specify
11176 respectively the default (best suited) video and audio stream. Default
11177 is "dv", or "da" if the filter is called as "amovie".
11179 @item stream_index, si
11180 Specifies the index of the video stream to read. If the value is -1,
11181 the most suitable video stream will be automatically selected. The default
11182 value is "-1". Deprecated. If the filter is called "amovie", it will select
11183 audio instead of video.
11186 Specifies how many times to read the stream in sequence.
11187 If the value is less than 1, the stream will be read again and again.
11188 Default value is "1".
11190 Note that when the movie is looped the source timestamps are not
11191 changed, so it will generate non monotonically increasing timestamps.
11194 It allows overlaying a second video on top of the main input of
11195 a filtergraph, as shown in this graph:
11197 input -----------> deltapts0 --> overlay --> output
11200 movie --> scale--> deltapts1 -------+
11202 @subsection Examples
11206 Skip 3.2 seconds from the start of the AVI file in.avi, and overlay it
11207 on top of the input labelled "in":
11209 movie=in.avi:seek_point=3.2, scale=180:-1, setpts=PTS-STARTPTS [over];
11210 [in] setpts=PTS-STARTPTS [main];
11211 [main][over] overlay=16:16 [out]
11215 Read from a video4linux2 device, and overlay it on top of the input
11218 movie=/dev/video0:f=video4linux2, scale=180:-1, setpts=PTS-STARTPTS [over];
11219 [in] setpts=PTS-STARTPTS [main];
11220 [main][over] overlay=16:16 [out]
11224 Read the first video stream and the audio stream with id 0x81 from
11225 dvd.vob; the video is connected to the pad named "video" and the audio is
11226 connected to the pad named "audio":
11228 movie=dvd.vob:s=v:0+#0x81 [video] [audio]
11232 @c man end MULTIMEDIA SOURCES