1 @chapter Filtering Introduction
2 @c man begin FILTERING INTRODUCTION
4 Filtering in FFmpeg is enabled through the libavfilter library.
6 Libavfilter is the filtering API of FFmpeg. It is the substitute of
7 the now deprecated 'vhooks' and started as a Google Summer of Code
10 Audio filtering integration into the main FFmpeg repository is a work in
11 progress, so audio API and ABI should not be considered stable yet.
13 In libavfilter, it is possible for filters to have multiple inputs and
15 To illustrate the sorts of things that are possible, we can
16 use a complex filter graph. For example, the following one:
19 input --> split --> fifo -----------------------> overlay --> output
22 +------> fifo --> crop --> vflip --------+
25 splits the stream in two streams, sends one stream through the crop filter
26 and the vflip filter before merging it back with the other stream by
27 overlaying it on top. You can use the following command to achieve this:
30 ffmpeg -i input -vf "[in] split [T1], fifo, [T2] overlay=0:H/2 [out]; [T1] fifo, crop=iw:ih/2:0:ih/2, vflip [T2]" output
33 The result will be that in output the top half of the video is mirrored
36 Filters are loaded using the @var{-vf} or @var{-af} option passed to
37 @command{ffmpeg} or to @command{ffplay}. Filters in the same linear
38 chain are separated by commas. In our example, @var{split, fifo,
39 overlay} are in one linear chain, and @var{fifo, crop, vflip} are in
40 another. The points where the linear chains join are labeled by names
41 enclosed in square brackets. In our example, that is @var{[T1]} and
42 @var{[T2]}. The special labels @var{[in]} and @var{[out]} are the points
43 where video is input and output.
45 Some filters take in input a list of parameters: they are specified
46 after the filter name and an equal sign, and are separated from each other
49 There exist so-called @var{source filters} that do not have an
50 audio/video input, and @var{sink filters} that will not have audio/video
53 @c man end FILTERING INTRODUCTION
56 @c man begin GRAPH2DOT
58 The @file{graph2dot} program included in the FFmpeg @file{tools}
59 directory can be used to parse a filter graph description and issue a
60 corresponding textual representation in the dot language.
67 to see how to use @file{graph2dot}.
69 You can then pass the dot description to the @file{dot} program (from
70 the graphviz suite of programs) and obtain a graphical representation
73 For example the sequence of commands:
75 echo @var{GRAPH_DESCRIPTION} | \
76 tools/graph2dot -o graph.tmp && \
77 dot -Tpng graph.tmp -o graph.png && \
81 can be used to create and display an image representing the graph
82 described by the @var{GRAPH_DESCRIPTION} string. Note that this string must be
83 a complete self-contained graph, with its inputs and outputs explicitly defined.
84 For example if your command line is of the form:
86 ffmpeg -i infile -vf scale=640:360 outfile
88 your @var{GRAPH_DESCRIPTION} string will need to be of the form:
90 nullsrc,scale=640:360,nullsink
92 you may also need to set the @var{nullsrc} parameters and add a @var{format}
93 filter in order to simulate a specific input file.
97 @chapter Filtergraph description
98 @c man begin FILTERGRAPH DESCRIPTION
100 A filtergraph is a directed graph of connected filters. It can contain
101 cycles, and there can be multiple links between a pair of
102 filters. Each link has one input pad on one side connecting it to one
103 filter from which it takes its input, and one output pad on the other
104 side connecting it to the one filter accepting its output.
106 Each filter in a filtergraph is an instance of a filter class
107 registered in the application, which defines the features and the
108 number of input and output pads of the filter.
110 A filter with no input pads is called a "source", a filter with no
111 output pads is called a "sink".
113 @anchor{Filtergraph syntax}
114 @section Filtergraph syntax
116 A filtergraph can be represented using a textual representation, which is
117 recognized by the @option{-filter}/@option{-vf} and @option{-filter_complex}
118 options in @command{ffmpeg} and @option{-vf} in @command{ffplay}, and by the
119 @code{avfilter_graph_parse()}/@code{avfilter_graph_parse2()} function defined in
120 @file{libavfilter/avfiltergraph.h}.
122 A filterchain consists of a sequence of connected filters, each one
123 connected to the previous one in the sequence. A filterchain is
124 represented by a list of ","-separated filter descriptions.
126 A filtergraph consists of a sequence of filterchains. A sequence of
127 filterchains is represented by a list of ";"-separated filterchain
130 A filter is represented by a string of the form:
131 [@var{in_link_1}]...[@var{in_link_N}]@var{filter_name}=@var{arguments}[@var{out_link_1}]...[@var{out_link_M}]
133 @var{filter_name} is the name of the filter class of which the
134 described filter is an instance of, and has to be the name of one of
135 the filter classes registered in the program.
136 The name of the filter class is optionally followed by a string
139 @var{arguments} is a string which contains the parameters used to
140 initialize the filter instance, and are described in the filter
143 The list of arguments can be quoted using the character "'" as initial
144 and ending mark, and the character '\' for escaping the characters
145 within the quoted text; otherwise the argument string is considered
146 terminated when the next special character (belonging to the set
147 "[]=;,") is encountered.
149 The name and arguments of the filter are optionally preceded and
150 followed by a list of link labels.
151 A link label allows to name a link and associate it to a filter output
152 or input pad. The preceding labels @var{in_link_1}
153 ... @var{in_link_N}, are associated to the filter input pads,
154 the following labels @var{out_link_1} ... @var{out_link_M}, are
155 associated to the output pads.
157 When two link labels with the same name are found in the
158 filtergraph, a link between the corresponding input and output pad is
161 If an output pad is not labelled, it is linked by default to the first
162 unlabelled input pad of the next filter in the filterchain.
163 For example in the filterchain:
165 nullsrc, split[L1], [L2]overlay, nullsink
167 the split filter instance has two output pads, and the overlay filter
168 instance two input pads. The first output pad of split is labelled
169 "L1", the first input pad of overlay is labelled "L2", and the second
170 output pad of split is linked to the second input pad of overlay,
171 which are both unlabelled.
173 In a complete filterchain all the unlabelled filter input and output
174 pads must be connected. A filtergraph is considered valid if all the
175 filter input and output pads of all the filterchains are connected.
177 Libavfilter will automatically insert scale filters where format
178 conversion is required. It is possible to specify swscale flags
179 for those automatically inserted scalers by prepending
180 @code{sws_flags=@var{flags};}
181 to the filtergraph description.
183 Follows a BNF description for the filtergraph syntax:
185 @var{NAME} ::= sequence of alphanumeric characters and '_'
186 @var{LINKLABEL} ::= "[" @var{NAME} "]"
187 @var{LINKLABELS} ::= @var{LINKLABEL} [@var{LINKLABELS}]
188 @var{FILTER_ARGUMENTS} ::= sequence of chars (eventually quoted)
189 @var{FILTER} ::= [@var{LINKNAMES}] @var{NAME} ["=" @var{ARGUMENTS}] [@var{LINKNAMES}]
190 @var{FILTERCHAIN} ::= @var{FILTER} [,@var{FILTERCHAIN}]
191 @var{FILTERGRAPH} ::= [sws_flags=@var{flags};] @var{FILTERCHAIN} [;@var{FILTERGRAPH}]
194 @section Notes on filtergraph escaping
196 Some filter arguments require the use of special characters, typically
197 @code{:} to separate key=value pairs in a named options list. In this
198 case the user should perform a first level escaping when specifying
199 the filter arguments. For example, consider the following literal
200 string to be embedded in the @ref{drawtext} filter arguments:
202 this is a 'string': may contain one, or more, special characters
205 Since @code{:} is special for the filter arguments syntax, it needs to
206 be escaped, so you get:
208 text=this is a \'string\'\: may contain one, or more, special characters
211 A second level of escaping is required when embedding the filter
212 arguments in a filtergraph description, in order to escape all the
213 filtergraph special characters. Thus the example above becomes:
215 drawtext=text=this is a \\\'string\\\'\\: may contain one\, or more\, special characters
218 Finally an additional level of escaping may be needed when writing the
219 filtergraph description in a shell command, which depends on the
220 escaping rules of the adopted shell. For example, assuming that
221 @code{\} is special and needs to be escaped with another @code{\}, the
222 previous string will finally result in:
224 -vf "drawtext=text=this is a \\\\\\'string\\\\\\'\\\\: may contain one\\, or more\\, special characters"
227 Sometimes, it might be more convenient to employ quoting in place of
228 escaping. For example the string:
230 Caesar: tu quoque, Brute, fili mi
233 Can be quoted in the filter arguments as:
235 text='Caesar: tu quoque, Brute, fili mi'
238 And finally inserted in a filtergraph like:
240 drawtext=text=\'Caesar: tu quoque\, Brute\, fili mi\'
243 See the @ref{quoting_and_escaping, Quoting and escaping} section for
244 more information about the escaping and quoting rules adopted by
247 @c man end FILTERGRAPH DESCRIPTION
249 @chapter Audio Filters
250 @c man begin AUDIO FILTERS
252 When you configure your FFmpeg build, you can disable any of the
253 existing filters using @code{--disable-filters}.
254 The configure output will show the audio filters included in your
257 Below is a description of the currently available audio filters.
261 Convert the input audio format to the specified formats.
263 The filter accepts a string of the form:
264 "@var{sample_format}:@var{channel_layout}".
266 @var{sample_format} specifies the sample format, and can be a string or the
267 corresponding numeric value defined in @file{libavutil/samplefmt.h}. Use 'p'
268 suffix for a planar sample format.
270 @var{channel_layout} specifies the channel layout, and can be a string
271 or the corresponding number value defined in @file{libavutil/channel_layout.h}.
273 The special parameter "auto", signifies that the filter will
274 automatically select the output format depending on the output filter.
276 Some examples follow.
280 Convert input to float, planar, stereo:
286 Convert input to unsigned 8-bit, automatically select out channel layout:
294 Convert the input audio to one of the specified formats. The framework will
295 negotiate the most appropriate format to minimize conversions.
297 The filter accepts the following named parameters:
301 A comma-separated list of requested sample formats.
304 A comma-separated list of requested sample rates.
306 @item channel_layouts
307 A comma-separated list of requested channel layouts.
311 If a parameter is omitted, all values are allowed.
313 For example to force the output to either unsigned 8-bit or signed 16-bit stereo:
315 aformat=sample_fmts\=u8\,s16:channel_layouts\=stereo
320 Merge two or more audio streams into a single multi-channel stream.
322 The filter accepts the following named options:
327 Set the number of inputs. Default is 2.
331 If the channel layouts of the inputs are disjoint, and therefore compatible,
332 the channel layout of the output will be set accordingly and the channels
333 will be reordered as necessary. If the channel layouts of the inputs are not
334 disjoint, the output will have all the channels of the first input then all
335 the channels of the second input, in that order, and the channel layout of
336 the output will be the default value corresponding to the total number of
339 For example, if the first input is in 2.1 (FL+FR+LF) and the second input
340 is FC+BL+BR, then the output will be in 5.1, with the channels in the
341 following order: a1, a2, b1, a3, b2, b3 (a1 is the first channel of the
342 first input, b1 is the first channel of the second input).
344 On the other hand, if both input are in stereo, the output channels will be
345 in the default order: a1, a2, b1, b2, and the channel layout will be
346 arbitrarily set to 4.0, which may or may not be the expected value.
348 All inputs must have the same sample rate, and format.
350 If inputs do not have the same duration, the output will stop with the
353 Example: merge two mono files into a stereo stream:
355 amovie=left.wav [l] ; amovie=right.mp3 [r] ; [l] [r] amerge
358 Example: multiple merges:
361 amovie=input.mkv:si=0 [a0];
362 amovie=input.mkv:si=1 [a1];
363 amovie=input.mkv:si=2 [a2];
364 amovie=input.mkv:si=3 [a3];
365 amovie=input.mkv:si=4 [a4];
366 amovie=input.mkv:si=5 [a5];
367 [a0][a1][a2][a3][a4][a5] amerge=inputs=6" -c:a pcm_s16le output.mkv
372 Mixes multiple audio inputs into a single output.
376 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex amix=inputs=3:duration=first:dropout_transition=3 OUTPUT
378 will mix 3 input audio streams to a single output with the same duration as the
379 first input and a dropout transition time of 3 seconds.
381 The filter accepts the following named parameters:
385 Number of inputs. If unspecified, it defaults to 2.
388 How to determine the end-of-stream.
392 Duration of longest input. (default)
395 Duration of shortest input.
398 Duration of first input.
402 @item dropout_transition
403 Transition time, in seconds, for volume renormalization when an input
404 stream ends. The default value is 2 seconds.
410 Pass the audio source unchanged to the output.
414 Resample the input audio to the specified sample rate.
416 The filter accepts exactly one parameter, the output sample rate. If not
417 specified then the filter will automatically convert between its input
418 and output sample rates.
420 For example, to resample the input audio to 44100Hz:
425 @section asetnsamples
427 Set the number of samples per each output audio frame.
429 The last output packet may contain a different number of samples, as
430 the filter will flush all the remaining samples when the input audio
433 The filter accepts parameters as a list of @var{key}=@var{value} pairs,
438 @item nb_out_samples, n
439 Set the number of frames per each output audio frame. The number is
440 intended as the number of samples @emph{per each channel}.
441 Default value is 1024.
444 If set to 1, the filter will pad the last audio frame with zeroes, so
445 that the last frame will contain the same number of samples as the
446 previous ones. Default value is 1.
449 For example, to set the number of per-frame samples to 1234 and
450 disable padding for the last frame, use:
452 asetnsamples=n=1234:p=0
457 Show a line containing various information for each input audio frame.
458 The input audio is not modified.
460 The shown line contains a sequence of key/value pairs of the form
461 @var{key}:@var{value}.
463 A description of each shown parameter follows:
467 sequential number of the input frame, starting from 0
470 Presentation timestamp of the input frame, in time base units; the time base
471 depends on the filter input pad, and is usually 1/@var{sample_rate}.
474 presentation timestamp of the input frame in seconds
477 position of the frame in the input stream, -1 if this information in
478 unavailable and/or meaningless (for example in case of synthetic audio)
487 sample rate for the audio frame
490 number of samples (per channel) in the frame
493 Adler-32 checksum (printed in hexadecimal) of the audio data. For planar audio
494 the data is treated as if all the planes were concatenated.
496 @item plane_checksums
497 A list of Adler-32 checksums for each data plane.
502 Split input audio into several identical outputs.
504 The filter accepts a single parameter which specifies the number of outputs. If
505 unspecified, it defaults to 2.
509 [in] asplit [out0][out1]
512 will create two separate outputs from the same input.
514 To create 3 or more outputs, you need to specify the number of
517 [in] asplit=3 [out0][out1][out2]
521 ffmpeg -i INPUT -filter_complex asplit=5 OUTPUT
523 will create 5 copies of the input audio.
528 Forward two audio streams and control the order the buffers are forwarded.
530 The argument to the filter is an expression deciding which stream should be
531 forwarded next: if the result is negative, the first stream is forwarded; if
532 the result is positive or zero, the second stream is forwarded. It can use
533 the following variables:
537 number of buffers forwarded so far on each stream
539 number of samples forwarded so far on each stream
541 current timestamp of each stream
544 The default value is @code{t1-t2}, which means to always forward the stream
545 that has a smaller timestamp.
547 Example: stress-test @code{amerge} by randomly sending buffers on the wrong
548 input, while avoiding too much of a desynchronization:
550 amovie=file.ogg [a] ; amovie=file.mp3 [b] ;
551 [a] [b] astreamsync=(2*random(1))-1+tanh(5*(t1-t2)) [a2] [b2] ;
559 The filter accepts exactly one parameter, the audio tempo. If not
560 specified then the filter will assume nominal 1.0 tempo. Tempo must
561 be in the [0.5, 2.0] range.
563 For example, to slow down audio to 80% tempo:
568 For example, to speed up audio to 125% tempo:
575 Make audio easier to listen to on headphones.
577 This filter adds `cues' to 44.1kHz stereo (i.e. audio CD format) audio
578 so that when listened to on headphones the stereo image is moved from
579 inside your head (standard for headphones) to outside and in front of
580 the listener (standard for speakers).
586 Mix channels with specific gain levels. The filter accepts the output
587 channel layout followed by a set of channels definitions.
589 This filter is also designed to remap efficiently the channels of an audio
592 The filter accepts parameters of the form:
593 "@var{l}:@var{outdef}:@var{outdef}:..."
597 output channel layout or number of channels
600 output channel specification, of the form:
601 "@var{out_name}=[@var{gain}*]@var{in_name}[+[@var{gain}*]@var{in_name}...]"
604 output channel to define, either a channel name (FL, FR, etc.) or a channel
605 number (c0, c1, etc.)
608 multiplicative coefficient for the channel, 1 leaving the volume unchanged
611 input channel to use, see out_name for details; it is not possible to mix
612 named and numbered input channels
615 If the `=' in a channel specification is replaced by `<', then the gains for
616 that specification will be renormalized so that the total is 1, thus
617 avoiding clipping noise.
619 @subsection Mixing examples
621 For example, if you want to down-mix from stereo to mono, but with a bigger
622 factor for the left channel:
624 pan=1:c0=0.9*c0+0.1*c1
627 A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
630 pan=stereo: FL < FL + 0.5*FC + 0.6*BL + 0.6*SL : FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
633 Note that @command{ffmpeg} integrates a default down-mix (and up-mix) system
634 that should be preferred (see "-ac" option) unless you have very specific
637 @subsection Remapping examples
639 The channel remapping will be effective if, and only if:
642 @item gain coefficients are zeroes or ones,
643 @item only one input per channel output,
646 If all these conditions are satisfied, the filter will notify the user ("Pure
647 channel mapping detected"), and use an optimized and lossless method to do the
650 For example, if you have a 5.1 source and want a stereo audio stream by
651 dropping the extra channels:
653 pan="stereo: c0=FL : c1=FR"
656 Given the same source, you can also switch front left and front right channels
657 and keep the input channel layout:
659 pan="5.1: c0=c1 : c1=c0 : c2=c2 : c3=c3 : c4=c4 : c5=c5"
662 If the input is a stereo audio stream, you can mute the front left channel (and
663 still keep the stereo channel layout) with:
668 Still with a stereo audio stream input, you can copy the right channel in both
669 front left and right:
671 pan="stereo: c0=FR : c1=FR"
674 @section silencedetect
676 Detect silence in an audio stream.
678 This filter logs a message when it detects that the input audio volume is less
679 or equal to a noise tolerance value for a duration greater or equal to the
680 minimum detected noise duration.
682 The printed times and duration are expressed in seconds.
686 Set silence duration until notification (default is 2 seconds).
689 Set noise tolerance. Can be specified in dB (in case "dB" is appended to the
690 specified value) or amplitude ratio. Default is -60dB, or 0.001.
693 Detect 5 seconds of silence with -50dB noise tolerance:
695 silencedetect=n=-50dB:d=5
698 Complete example with @command{ffmpeg} to detect silence with 0.0001 noise
699 tolerance in @file{silence.mp3}:
701 ffmpeg -f lavfi -i amovie=silence.mp3,silencedetect=noise=0.0001 -f null -
706 Adjust the input audio volume.
708 The filter accepts exactly one parameter @var{vol}, which expresses
709 how the audio volume will be increased or decreased.
711 Output values are clipped to the maximum value.
713 If @var{vol} is expressed as a decimal number, the output audio
714 volume is given by the relation:
716 @var{output_volume} = @var{vol} * @var{input_volume}
719 If @var{vol} is expressed as a decimal number followed by the string
720 "dB", the value represents the requested change in decibels of the
721 input audio power, and the output audio volume is given by the
724 @var{output_volume} = 10^(@var{vol}/20) * @var{input_volume}
727 Otherwise @var{vol} is considered an expression and its evaluated
728 value is used for computing the output audio volume according to the
731 Default value for @var{vol} is 1.0.
737 Half the input audio volume:
742 The above example is equivalent to:
748 Decrease input audio power by 12 decibels:
754 @section volumedetect
756 Detect the volume of the input video.
758 The filter has no parameters. The input is not modified. Statistics about
759 the volume will be printed in the log when the input stream end is reached.
761 In particular it will show the mean volume (root mean square), maximum
762 volume (on a per-sample basis), and the beginning of an histogram of the
763 registered volume values (from the maximum value to a cumulated 1/1000 of
766 All volumes are in decibels relative to the maximum PCM value.
768 Here is an excerpt of the output:
770 [Parsed_volumedetect_0 @ 0xa23120] mean_volume: -27 dB
771 [Parsed_volumedetect_0 @ 0xa23120] max_volume: -4 dB
772 [Parsed_volumedetect_0 @ 0xa23120] histogram_4db: 6
773 [Parsed_volumedetect_0 @ 0xa23120] histogram_5db: 62
774 [Parsed_volumedetect_0 @ 0xa23120] histogram_6db: 286
775 [Parsed_volumedetect_0 @ 0xa23120] histogram_7db: 1042
776 [Parsed_volumedetect_0 @ 0xa23120] histogram_8db: 2551
777 [Parsed_volumedetect_0 @ 0xa23120] histogram_9db: 4609
778 [Parsed_volumedetect_0 @ 0xa23120] histogram_10db: 8409
784 The mean square energy is approximately -27 dB, or 10^-2.7.
786 The largest sample is at -4 dB, or more precisely between -4 dB and -5 dB.
788 There are 6 samples at -4 dB, 62 at -5 dB, 286 at -6 dB, etc.
791 In other words, raising the volume by +4 dB does not cause any clipping,
792 raising it by +5 dB causes clipping for 6 samples, etc.
795 Synchronize audio data with timestamps by squeezing/stretching it and/or
796 dropping samples/adding silence when needed.
798 The filter accepts the following named parameters:
802 Enable stretching/squeezing the data to make it match the timestamps. Disabled
803 by default. When disabled, time gaps are covered with silence.
806 Minimum difference between timestamps and audio data (in seconds) to trigger
807 adding/dropping samples. Default value is 0.1. If you get non-perfect sync with
808 this filter, try setting this parameter to 0.
811 Maximum compensation in samples per second. Relevant only with compensate=1.
815 Assume the first pts should be this value.
816 This allows for padding/trimming at the start of stream. By default, no
817 assumption is made about the first frame's expected pts, so no padding or
818 trimming is done. For example, this could be set to 0 to pad the beginning with
819 silence if an audio stream starts after the video stream.
823 @section channelsplit
824 Split each channel in input audio stream into a separate output stream.
826 This filter accepts the following named parameters:
829 Channel layout of the input stream. Default is "stereo".
832 For example, assuming a stereo input MP3 file
834 ffmpeg -i in.mp3 -filter_complex channelsplit out.mkv
836 will create an output Matroska file with two audio streams, one containing only
837 the left channel and the other the right channel.
839 To split a 5.1 WAV file into per-channel files
841 ffmpeg -i in.wav -filter_complex
842 'channelsplit=channel_layout=5.1[FL][FR][FC][LFE][SL][SR]'
843 -map '[FL]' front_left.wav -map '[FR]' front_right.wav -map '[FC]'
844 front_center.wav -map '[LFE]' lfe.wav -map '[SL]' side_left.wav -map '[SR]'
849 Remap input channels to new locations.
851 This filter accepts the following named parameters:
854 Channel layout of the output stream.
857 Map channels from input to output. The argument is a comma-separated list of
858 mappings, each in the @code{@var{in_channel}-@var{out_channel}} or
859 @var{in_channel} form. @var{in_channel} can be either the name of the input
860 channel (e.g. FL for front left) or its index in the input channel layout.
861 @var{out_channel} is the name of the output channel or its index in the output
862 channel layout. If @var{out_channel} is not given then it is implicitly an
863 index, starting with zero and increasing by one for each mapping.
866 If no mapping is present, the filter will implicitly map input channels to
867 output channels preserving index.
869 For example, assuming a 5.1+downmix input MOV file
871 ffmpeg -i in.mov -filter 'channelmap=map=DL-FL\,DR-FR' out.wav
873 will create an output WAV file tagged as stereo from the downmix channels of
876 To fix a 5.1 WAV improperly encoded in AAC's native channel order
878 ffmpeg -i in.wav -filter 'channelmap=1\,2\,0\,5\,3\,4:channel_layout=5.1' out.wav
882 Join multiple input streams into one multi-channel stream.
884 The filter accepts the following named parameters:
888 Number of input streams. Defaults to 2.
891 Desired output channel layout. Defaults to stereo.
894 Map channels from inputs to output. The argument is a comma-separated list of
895 mappings, each in the @code{@var{input_idx}.@var{in_channel}-@var{out_channel}}
896 form. @var{input_idx} is the 0-based index of the input stream. @var{in_channel}
897 can be either the name of the input channel (e.g. FL for front left) or its
898 index in the specified input stream. @var{out_channel} is the name of the output
902 The filter will attempt to guess the mappings when those are not specified
903 explicitly. It does so by first trying to find an unused matching input channel
904 and if that fails it picks the first unused input channel.
906 E.g. to join 3 inputs (with properly set channel layouts)
908 ffmpeg -i INPUT1 -i INPUT2 -i INPUT3 -filter_complex join=inputs=3 OUTPUT
911 To build a 5.1 output from 6 single-channel streams:
913 ffmpeg -i fl -i fr -i fc -i sl -i sr -i lfe -filter_complex
914 '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'
919 Convert the audio sample format, sample rate and channel layout. This filter is
920 not meant to be used directly.
922 @c man end AUDIO FILTERS
924 @chapter Audio Sources
925 @c man begin AUDIO SOURCES
927 Below is a description of the currently available audio sources.
931 Buffer audio frames, and make them available to the filter chain.
933 This source is mainly intended for a programmatic use, in particular
934 through the interface defined in @file{libavfilter/asrc_abuffer.h}.
936 It accepts the following mandatory parameters:
937 @var{sample_rate}:@var{sample_fmt}:@var{channel_layout}
942 The sample rate of the incoming audio buffers.
945 The sample format of the incoming audio buffers.
946 Either a sample format name or its corresponging integer representation from
947 the enum AVSampleFormat in @file{libavutil/samplefmt.h}
950 The channel layout of the incoming audio buffers.
951 Either a channel layout name from channel_layout_map in
952 @file{libavutil/channel_layout.c} or its corresponding integer representation
953 from the AV_CH_LAYOUT_* macros in @file{libavutil/channel_layout.h}
959 abuffer=44100:s16p:stereo
962 will instruct the source to accept planar 16bit signed stereo at 44100Hz.
963 Since the sample format with name "s16p" corresponds to the number
964 6 and the "stereo" channel layout corresponds to the value 0x3, this is
972 Generate an audio signal specified by an expression.
974 This source accepts in input one or more expressions (one for each
975 channel), which are evaluated and used to generate a corresponding
978 It accepts the syntax: @var{exprs}[::@var{options}].
979 @var{exprs} is a list of expressions separated by ":", one for each
980 separate channel. In case the @var{channel_layout} is not
981 specified, the selected channel layout depends on the number of
982 provided expressions.
984 @var{options} is an optional sequence of @var{key}=@var{value} pairs,
987 The description of the accepted options follows.
991 @item channel_layout, c
992 Set the channel layout. The number of channels in the specified layout
993 must be equal to the number of specified expressions.
996 Set the minimum duration of the sourced audio. See the function
997 @code{av_parse_time()} for the accepted format.
998 Note that the resulting duration may be greater than the specified
999 duration, as the generated audio is always cut at the end of a
1002 If not specified, or the expressed duration is negative, the audio is
1003 supposed to be generated forever.
1006 Set the number of samples per channel per each output frame,
1009 @item sample_rate, s
1010 Specify the sample rate, default to 44100.
1013 Each expression in @var{exprs} can contain the following constants:
1017 number of the evaluated sample, starting from 0
1020 time of the evaluated sample expressed in seconds, starting from 0
1027 @subsection Examples
1039 Generate a sin signal with frequency of 440 Hz, set sample rate to
1042 aevalsrc="sin(440*2*PI*t)::s=8000"
1046 Generate a two channels signal, specify the channel layout (Front
1047 Center + Back Center) explicitly:
1049 aevalsrc="sin(420*2*PI*t):cos(430*2*PI*t)::c=FC|BC"
1053 Generate white noise:
1055 aevalsrc="-2+random(0)"
1059 Generate an amplitude modulated signal:
1061 aevalsrc="sin(10*2*PI*t)*sin(880*2*PI*t)"
1065 Generate 2.5 Hz binaural beats on a 360 Hz carrier:
1067 aevalsrc="0.1*sin(2*PI*(360-2.5/2)*t) : 0.1*sin(2*PI*(360+2.5/2)*t)"
1074 Null audio source, return unprocessed audio frames. It is mainly useful
1075 as a template and to be employed in analysis / debugging tools, or as
1076 the source for filters which ignore the input data (for example the sox
1079 It accepts an optional sequence of @var{key}=@var{value} pairs,
1082 The description of the accepted options follows.
1086 @item sample_rate, s
1087 Specify the sample rate, and defaults to 44100.
1089 @item channel_layout, cl
1091 Specify the channel layout, and can be either an integer or a string
1092 representing a channel layout. The default value of @var{channel_layout}
1095 Check the channel_layout_map definition in
1096 @file{libavutil/channel_layout.c} for the mapping between strings and
1097 channel layout values.
1100 Set the number of samples per requested frames.
1104 Follow some examples:
1106 # set the sample rate to 48000 Hz and the channel layout to AV_CH_LAYOUT_MONO.
1107 anullsrc=r=48000:cl=4
1110 anullsrc=r=48000:cl=mono
1114 Buffer audio frames, and make them available to the filter chain.
1116 This source is not intended to be part of user-supplied graph descriptions but
1117 for insertion by calling programs through the interface defined in
1118 @file{libavfilter/buffersrc.h}.
1120 It accepts the following named parameters:
1124 Timebase which will be used for timestamps of submitted frames. It must be
1125 either a floating-point number or in @var{numerator}/@var{denominator} form.
1131 Name of the sample format, as returned by @code{av_get_sample_fmt_name()}.
1133 @item channel_layout
1134 Channel layout of the audio data, in the form that can be accepted by
1135 @code{av_get_channel_layout()}.
1138 All the parameters need to be explicitly defined.
1142 Synthesize a voice utterance using the libflite library.
1144 To enable compilation of this filter you need to configure FFmpeg with
1145 @code{--enable-libflite}.
1147 Note that the flite library is not thread-safe.
1149 The source accepts parameters as a list of @var{key}=@var{value} pairs,
1152 The description of the accepted parameters follows.
1157 If set to 1, list the names of the available voices and exit
1158 immediately. Default value is 0.
1161 Set the maximum number of samples per frame. Default value is 512.
1164 Set the filename containing the text to speak.
1167 Set the text to speak.
1170 Set the voice to use for the speech synthesis. Default value is
1171 @code{kal}. See also the @var{list_voices} option.
1174 @subsection Examples
1178 Read from file @file{speech.txt}, and synthetize the text using the
1179 standard flite voice:
1181 flite=textfile=speech.txt
1185 Read the specified text selecting the @code{slt} voice:
1187 flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
1191 Input text to ffmpeg:
1193 ffmpeg -f lavfi -i flite=text='So fare thee well, poor devil of a Sub-Sub, whose commentator I am':voice=slt
1197 Make @file{ffplay} speak the specified text, using @code{flite} and
1198 the @code{lavfi} device:
1200 ffplay -f lavfi flite=text='No more be grieved for which that thou hast done.'
1204 For more information about libflite, check:
1205 @url{http://www.speech.cs.cmu.edu/flite/}
1207 @c man end AUDIO SOURCES
1209 @chapter Audio Sinks
1210 @c man begin AUDIO SINKS
1212 Below is a description of the currently available audio sinks.
1214 @section abuffersink
1216 Buffer audio frames, and make them available to the end of filter chain.
1218 This sink is mainly intended for programmatic use, in particular
1219 through the interface defined in @file{libavfilter/buffersink.h}.
1221 It requires a pointer to an AVABufferSinkContext structure, which
1222 defines the incoming buffers' formats, to be passed as the opaque
1223 parameter to @code{avfilter_init_filter} for initialization.
1227 Null audio sink, do absolutely nothing with the input audio. It is
1228 mainly useful as a template and to be employed in analysis / debugging
1231 @section abuffersink
1232 This sink is intended for programmatic use. Frames that arrive on this sink can
1233 be retrieved by the calling program using the interface defined in
1234 @file{libavfilter/buffersink.h}.
1236 This filter accepts no parameters.
1238 @c man end AUDIO SINKS
1240 @chapter Video Filters
1241 @c man begin VIDEO FILTERS
1243 When you configure your FFmpeg build, you can disable any of the
1244 existing filters using @code{--disable-filters}.
1245 The configure output will show the video filters included in your
1248 Below is a description of the currently available video filters.
1250 @section alphaextract
1252 Extract the alpha component from the input as a grayscale video. This
1253 is especially useful with the @var{alphamerge} filter.
1257 Add or replace the alpha component of the primary input with the
1258 grayscale value of a second input. This is intended for use with
1259 @var{alphaextract} to allow the transmission or storage of frame
1260 sequences that have alpha in a format that doesn't support an alpha
1263 For example, to reconstruct full frames from a normal YUV-encoded video
1264 and a separate video created with @var{alphaextract}, you might use:
1266 movie=in_alpha.mkv [alpha]; [in][alpha] alphamerge [out]
1269 Since this filter is designed for reconstruction, it operates on frame
1270 sequences without considering timestamps, and terminates when either
1271 input reaches end of stream. This will cause problems if your encoding
1272 pipeline drops frames. If you're trying to apply an image as an
1273 overlay to a video stream, consider the @var{overlay} filter instead.
1277 Draw ASS (Advanced Substation Alpha) subtitles on top of input video
1278 using the libass library.
1280 To enable compilation of this filter you need to configure FFmpeg with
1281 @code{--enable-libass}.
1283 This filter accepts the following named options, expressed as a
1284 sequence of @var{key}=@var{value} pairs, separated by ":".
1288 Set the filename of the ASS file to read. It must be specified.
1291 Specify the size of the original video, the video for which the ASS file
1292 was composed. Due to a misdesign in ASS aspect ratio arithmetic, this is
1293 necessary to correctly scale the fonts if the aspect ratio has been changed.
1296 If the first key is not specified, it is assumed that the first value
1297 specifies the @option{filename}.
1299 For example, to render the file @file{sub.ass} on top of the input
1300 video, use the command:
1305 which is equivalent to:
1307 ass=filename=sub.ass
1312 Compute the bounding box for the non-black pixels in the input frame
1315 This filter computes the bounding box containing all the pixels with a
1316 luminance value greater than the minimum allowed value.
1317 The parameters describing the bounding box are printed on the filter
1320 @section blackdetect
1322 Detect video intervals that are (almost) completely black. Can be
1323 useful to detect chapter transitions, commercials, or invalid
1324 recordings. Output lines contains the time for the start, end and
1325 duration of the detected black interval expressed in seconds.
1327 In order to display the output lines, you need to set the loglevel at
1328 least to the AV_LOG_INFO value.
1330 This filter accepts a list of options in the form of
1331 @var{key}=@var{value} pairs separated by ":". A description of the
1332 accepted options follows.
1335 @item black_min_duration, d
1336 Set the minimum detected black duration expressed in seconds. It must
1337 be a non-negative floating point number.
1339 Default value is 2.0.
1341 @item picture_black_ratio_th, pic_th
1342 Set the threshold for considering a picture "black".
1343 Express the minimum value for the ratio:
1345 @var{nb_black_pixels} / @var{nb_pixels}
1348 for which a picture is considered black.
1349 Default value is 0.98.
1351 @item pixel_black_th, pix_th
1352 Set the threshold for considering a pixel "black".
1354 The threshold expresses the maximum pixel luminance value for which a
1355 pixel is considered "black". The provided value is scaled according to
1356 the following equation:
1358 @var{absolute_threshold} = @var{luminance_minimum_value} + @var{pixel_black_th} * @var{luminance_range_size}
1361 @var{luminance_range_size} and @var{luminance_minimum_value} depend on
1362 the input video format, the range is [0-255] for YUV full-range
1363 formats and [16-235] for YUV non full-range formats.
1365 Default value is 0.10.
1368 The following example sets the maximum pixel threshold to the minimum
1369 value, and detects only black intervals of 2 or more seconds:
1371 blackdetect=d=2:pix_th=0.00
1376 Detect frames that are (almost) completely black. Can be useful to
1377 detect chapter transitions or commercials. Output lines consist of
1378 the frame number of the detected frame, the percentage of blackness,
1379 the position in the file if known or -1 and the timestamp in seconds.
1381 In order to display the output lines, you need to set the loglevel at
1382 least to the AV_LOG_INFO value.
1384 The filter accepts the syntax:
1386 blackframe[=@var{amount}:[@var{threshold}]]
1389 @var{amount} is the percentage of the pixels that have to be below the
1390 threshold, and defaults to 98.
1392 @var{threshold} is the threshold below which a pixel value is
1393 considered black, and defaults to 32.
1397 Apply boxblur algorithm to the input video.
1399 This filter accepts the parameters:
1400 @var{luma_radius}:@var{luma_power}:@var{chroma_radius}:@var{chroma_power}:@var{alpha_radius}:@var{alpha_power}
1402 Chroma and alpha parameters are optional, if not specified they default
1403 to the corresponding values set for @var{luma_radius} and
1406 @var{luma_radius}, @var{chroma_radius}, and @var{alpha_radius} represent
1407 the radius in pixels of the box used for blurring the corresponding
1408 input plane. They are expressions, and can contain the following
1412 the input width and height in pixels
1415 the input chroma image width and height in pixels
1418 horizontal and vertical chroma subsample values. For example for the
1419 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
1422 The radius must be a non-negative number, and must not be greater than
1423 the value of the expression @code{min(w,h)/2} for the luma and alpha planes,
1424 and of @code{min(cw,ch)/2} for the chroma planes.
1426 @var{luma_power}, @var{chroma_power}, and @var{alpha_power} represent
1427 how many times the boxblur filter is applied to the corresponding
1430 Some examples follow:
1435 Apply a boxblur filter with luma, chroma, and alpha radius
1442 Set luma radius to 2, alpha and chroma radius to 0
1448 Set luma and chroma radius to a fraction of the video dimension
1450 boxblur=min(h\,w)/10:1:min(cw\,ch)/10:1
1455 @section colormatrix
1457 The colormatrix filter allows conversion between any of the following color
1458 space: BT.709 (@var{bt709}), BT.601 (@var{bt601}), SMPTE-240M (@var{smpte240m})
1459 and FCC (@var{fcc}).
1461 The syntax of the parameters is @var{source}:@var{destination}:
1464 colormatrix=bt601:smpte240m
1469 Copy the input source unchanged to the output. Mainly useful for
1474 Crop the input video to @var{out_w}:@var{out_h}:@var{x}:@var{y}:@var{keep_aspect}
1476 The @var{keep_aspect} parameter is optional, if specified and set to a
1477 non-zero value will force the output display aspect ratio to be the
1478 same of the input, by changing the output sample aspect ratio.
1480 The @var{out_w}, @var{out_h}, @var{x}, @var{y} parameters are
1481 expressions containing the following constants:
1485 the computed values for @var{x} and @var{y}. They are evaluated for
1489 the input width and height
1492 same as @var{in_w} and @var{in_h}
1495 the output (cropped) width and height
1498 same as @var{out_w} and @var{out_h}
1501 same as @var{iw} / @var{ih}
1504 input sample aspect ratio
1507 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
1510 horizontal and vertical chroma subsample values. For example for the
1511 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
1514 the number of input frame, starting from 0
1517 the position in the file of the input frame, NAN if unknown
1520 timestamp expressed in seconds, NAN if the input timestamp is unknown
1524 The @var{out_w} and @var{out_h} parameters specify the expressions for
1525 the width and height of the output (cropped) video. They are
1526 evaluated just at the configuration of the filter.
1528 The default value of @var{out_w} is "in_w", and the default value of
1529 @var{out_h} is "in_h".
1531 The expression for @var{out_w} may depend on the value of @var{out_h},
1532 and the expression for @var{out_h} may depend on @var{out_w}, but they
1533 cannot depend on @var{x} and @var{y}, as @var{x} and @var{y} are
1534 evaluated after @var{out_w} and @var{out_h}.
1536 The @var{x} and @var{y} parameters specify the expressions for the
1537 position of the top-left corner of the output (non-cropped) area. They
1538 are evaluated for each frame. If the evaluated value is not valid, it
1539 is approximated to the nearest valid value.
1541 The default value of @var{x} is "(in_w-out_w)/2", and the default
1542 value for @var{y} is "(in_h-out_h)/2", which set the cropped area at
1543 the center of the input image.
1545 The expression for @var{x} may depend on @var{y}, and the expression
1546 for @var{y} may depend on @var{x}.
1548 Follow some examples:
1550 # crop the central input area with size 100x100
1553 # crop the central input area with size 2/3 of the input video
1554 "crop=2/3*in_w:2/3*in_h"
1556 # crop the input video central square
1559 # delimit the rectangle with the top-left corner placed at position
1560 # 100:100 and the right-bottom corner corresponding to the right-bottom
1561 # corner of the input image.
1562 crop=in_w-100:in_h-100:100:100
1564 # crop 10 pixels from the left and right borders, and 20 pixels from
1565 # the top and bottom borders
1566 "crop=in_w-2*10:in_h-2*20"
1568 # keep only the bottom right quarter of the input image
1569 "crop=in_w/2:in_h/2:in_w/2:in_h/2"
1571 # crop height for getting Greek harmony
1572 "crop=in_w:1/PHI*in_w"
1575 "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)"
1577 # erratic camera effect depending on timestamp
1578 "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)"
1580 # set x depending on the value of y
1581 "crop=in_w/2:in_h/2:y:10+10*sin(n/10)"
1586 Auto-detect crop size.
1588 Calculate necessary cropping parameters and prints the recommended
1589 parameters through the logging system. The detected dimensions
1590 correspond to the non-black area of the input video.
1592 It accepts the syntax:
1594 cropdetect[=@var{limit}[:@var{round}[:@var{reset}]]]
1600 Threshold, which can be optionally specified from nothing (0) to
1601 everything (255), defaults to 24.
1604 Value which the width/height should be divisible by, defaults to
1605 16. The offset is automatically adjusted to center the video. Use 2 to
1606 get only even dimensions (needed for 4:2:2 video). 16 is best when
1607 encoding to most video codecs.
1610 Counter that determines after how many frames cropdetect will reset
1611 the previously detected largest video area and start over to detect
1612 the current optimal crop area. Defaults to 0.
1614 This can be useful when channel logos distort the video area. 0
1615 indicates never reset and return the largest area encountered during
1621 This filter drops frames that do not differ greatly from the previous
1622 frame in order to reduce framerate. The main use of this filter is
1623 for very-low-bitrate encoding (e.g. streaming over dialup modem), but
1624 it could in theory be used for fixing movies that were
1625 inverse-telecined incorrectly.
1627 It accepts the following parameters:
1628 @var{max}:@var{hi}:@var{lo}:@var{frac}.
1633 Set the maximum number of consecutive frames which can be dropped (if
1634 positive), or the minimum interval between dropped frames (if
1635 negative). If the value is 0, the frame is dropped unregarding the
1636 number of previous sequentially dropped frames.
1641 Set the dropping threshold values.
1643 Values for @var{hi} and @var{lo} are for 8x8 pixel blocks and
1644 represent actual pixel value differences, so a threshold of 64
1645 corresponds to 1 unit of difference for each pixel, or the same spread
1646 out differently over the block.
1648 A frame is a candidate for dropping if no 8x8 blocks differ by more
1649 than a threshold of @var{hi}, and if no more than @var{frac} blocks (1
1650 meaning the whole image) differ by more than a threshold of @var{lo}.
1652 Default value for @var{hi} is 64*12, default value for @var{lo} is
1653 64*5, and default value for @var{frac} is 0.33.
1658 Suppress a TV station logo by a simple interpolation of the surrounding
1659 pixels. Just set a rectangle covering the logo and watch it disappear
1660 (and sometimes something even uglier appear - your mileage may vary).
1662 The filter accepts parameters as a string of the form
1663 "@var{x}:@var{y}:@var{w}:@var{h}:@var{band}", or as a list of
1664 @var{key}=@var{value} pairs, separated by ":".
1666 The description of the accepted parameters follows.
1671 Specify the top left corner coordinates of the logo. They must be
1675 Specify the width and height of the logo to clear. They must be
1679 Specify the thickness of the fuzzy edge of the rectangle (added to
1680 @var{w} and @var{h}). The default value is 4.
1683 When set to 1, a green rectangle is drawn on the screen to simplify
1684 finding the right @var{x}, @var{y}, @var{w}, @var{h} parameters, and
1685 @var{band} is set to 4. The default value is 0.
1689 Some examples follow.
1694 Set a rectangle covering the area with top left corner coordinates 0,0
1695 and size 100x77, setting a band of size 10:
1697 delogo=0:0:100:77:10
1701 As the previous example, but use named options:
1703 delogo=x=0:y=0:w=100:h=77:band=10
1710 Attempt to fix small changes in horizontal and/or vertical shift. This
1711 filter helps remove camera shake from hand-holding a camera, bumping a
1712 tripod, moving on a vehicle, etc.
1714 The filter accepts parameters as a string of the form
1715 "@var{x}:@var{y}:@var{w}:@var{h}:@var{rx}:@var{ry}:@var{edge}:@var{blocksize}:@var{contrast}:@var{search}:@var{filename}"
1717 A description of the accepted parameters follows.
1722 Specify a rectangular area where to limit the search for motion
1724 If desired the search for motion vectors can be limited to a
1725 rectangular area of the frame defined by its top left corner, width
1726 and height. These parameters have the same meaning as the drawbox
1727 filter which can be used to visualise the position of the bounding
1730 This is useful when simultaneous movement of subjects within the frame
1731 might be confused for camera motion by the motion vector search.
1733 If any or all of @var{x}, @var{y}, @var{w} and @var{h} are set to -1
1734 then the full frame is used. This allows later options to be set
1735 without specifying the bounding box for the motion vector search.
1737 Default - search the whole frame.
1740 Specify the maximum extent of movement in x and y directions in the
1741 range 0-64 pixels. Default 16.
1744 Specify how to generate pixels to fill blanks at the edge of the
1745 frame. An integer from 0 to 3 as follows:
1748 Fill zeroes at blank locations
1750 Original image at blank locations
1752 Extruded edge value at blank locations
1754 Mirrored edge at blank locations
1757 The default setting is mirror edge at blank locations.
1760 Specify the blocksize to use for motion search. Range 4-128 pixels,
1764 Specify the contrast threshold for blocks. Only blocks with more than
1765 the specified contrast (difference between darkest and lightest
1766 pixels) will be considered. Range 1-255, default 125.
1769 Specify the search strategy 0 = exhaustive search, 1 = less exhaustive
1770 search. Default - exhaustive search.
1773 If set then a detailed log of the motion search is written to the
1780 Draw a colored box on the input image.
1782 The filter accepts parameters as a list of @var{key}=@var{value} pairs,
1785 The description of the accepted parameters follows.
1789 Specify the top left corner coordinates of the box. Default to 0.
1793 Specify the width and height of the box, if 0 they are interpreted as
1794 the input width and height. Default to 0.
1797 Specify the color of the box to write, it can be the name of a color
1798 (case insensitive match) or a 0xRRGGBB[AA] sequence. If the special
1799 value @code{invert} is used, the box edge color is the same as the
1800 video with inverted luma.
1803 Set the thickness of the box edge. Default value is @code{4}.
1806 If the key of the first options is omitted, the arguments are
1807 interpreted according to the following syntax:
1809 drawbox=@var{x}:@var{y}:@var{width}:@var{height}:@var{color}:@var{thickness}
1812 Some examples follow:
1815 Draw a black box around the edge of the input image:
1821 Draw a box with color red and an opacity of 50%:
1823 drawbox=10:20:200:60:red@@0.5
1826 The previous example can be specified as:
1828 drawbox=x=10:y=20:w=200:h=60:color=red@@0.5
1832 Fill the box with pink color:
1834 drawbox=x=10:y=10:w=100:h=100:color=pink@@0.5:t=max
1841 Draw text string or text from specified file on top of video using the
1842 libfreetype library.
1844 To enable compilation of this filter you need to configure FFmpeg with
1845 @code{--enable-libfreetype}.
1849 The filter accepts parameters as a list of @var{key}=@var{value} pairs,
1852 The description of the accepted parameters follows.
1857 Used to draw a box around text using background color.
1858 Value should be either 1 (enable) or 0 (disable).
1859 The default value of @var{box} is 0.
1862 The color to be used for drawing box around text.
1863 Either a string (e.g. "yellow") or in 0xRRGGBB[AA] format
1864 (e.g. "0xff00ff"), possibly followed by an alpha specifier.
1865 The default value of @var{boxcolor} is "white".
1868 Set an expression which specifies if the text should be drawn. If the
1869 expression evaluates to 0, the text is not drawn. This is useful for
1870 specifying that the text should be drawn only when specific conditions
1873 Default value is "1".
1875 See below for the list of accepted constants and functions.
1878 Select how the @var{text} is expanded. Can be either @code{none},
1879 @code{strftime} (default for compatibity reasons but deprecated) or
1880 @code{normal}. See the @ref{drawtext_expansion, Text expansion} section
1884 If true, check and fix text coords to avoid clipping.
1887 The color to be used for drawing fonts.
1888 Either a string (e.g. "red") or in 0xRRGGBB[AA] format
1889 (e.g. "0xff000033"), possibly followed by an alpha specifier.
1890 The default value of @var{fontcolor} is "black".
1893 The font file to be used for drawing text. Path must be included.
1894 This parameter is mandatory.
1897 The font size to be used for drawing text.
1898 The default value of @var{fontsize} is 16.
1901 Flags to be used for loading the fonts.
1903 The flags map the corresponding flags supported by libfreetype, and are
1904 a combination of the following values:
1911 @item vertical_layout
1912 @item force_autohint
1915 @item ignore_global_advance_width
1917 @item ignore_transform
1924 Default value is "render".
1926 For more information consult the documentation for the FT_LOAD_*
1930 The color to be used for drawing a shadow behind the drawn text. It
1931 can be a color name (e.g. "yellow") or a string in the 0xRRGGBB[AA]
1932 form (e.g. "0xff00ff"), possibly followed by an alpha specifier.
1933 The default value of @var{shadowcolor} is "black".
1935 @item shadowx, shadowy
1936 The x and y offsets for the text shadow position with respect to the
1937 position of the text. They can be either positive or negative
1938 values. Default value for both is "0".
1941 The size in number of spaces to use for rendering the tab.
1945 Set the initial timecode representation in "hh:mm:ss[:;.]ff"
1946 format. It can be used with or without text parameter. @var{timecode_rate}
1947 option must be specified.
1949 @item timecode_rate, rate, r
1950 Set the timecode frame rate (timecode only).
1953 The text string to be drawn. The text must be a sequence of UTF-8
1955 This parameter is mandatory if no file is specified with the parameter
1959 A text file containing text to be drawn. The text must be a sequence
1960 of UTF-8 encoded characters.
1962 This parameter is mandatory if no text string is specified with the
1963 parameter @var{text}.
1965 If both @var{text} and @var{textfile} are specified, an error is thrown.
1968 The expressions which specify the offsets where text will be drawn
1969 within the video frame. They are relative to the top/left border of the
1972 The default value of @var{x} and @var{y} is "0".
1974 See below for the list of accepted constants and functions.
1977 The parameters for @var{x} and @var{y} are expressions containing the
1978 following constants and functions:
1982 input display aspect ratio, it is the same as (@var{w} / @var{h}) * @var{sar}
1985 horizontal and vertical chroma subsample values. For example for the
1986 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
1989 the height of each text line
1997 @item max_glyph_a, ascent
1998 the maximum distance from the baseline to the highest/upper grid
1999 coordinate used to place a glyph outline point, for all the rendered
2001 It is a positive value, due to the grid's orientation with the Y axis
2004 @item max_glyph_d, descent
2005 the maximum distance from the baseline to the lowest grid coordinate
2006 used to place a glyph outline point, for all the rendered glyphs.
2007 This is a negative value, due to the grid's orientation, with the Y axis
2011 maximum glyph height, that is the maximum height for all the glyphs
2012 contained in the rendered text, it is equivalent to @var{ascent} -
2016 maximum glyph width, that is the maximum width for all the glyphs
2017 contained in the rendered text
2020 the number of input frame, starting from 0
2022 @item rand(min, max)
2023 return a random number included between @var{min} and @var{max}
2026 input sample aspect ratio
2029 timestamp expressed in seconds, NAN if the input timestamp is unknown
2032 the height of the rendered text
2035 the width of the rendered text
2038 the x and y offset coordinates where the text is drawn.
2040 These parameters allow the @var{x} and @var{y} expressions to refer
2041 each other, so you can for example specify @code{y=x/dar}.
2044 If libavfilter was built with @code{--enable-fontconfig}, then
2045 @option{fontfile} can be a fontconfig pattern or omitted.
2047 @anchor{drawtext_expansion}
2048 @subsection Text expansion
2050 If @option{expansion} is set to @code{strftime} (which is the default for
2051 now), the filter recognizes strftime() sequences in the provided text and
2052 expands them accordingly. Check the documentation of strftime(). This
2053 feature is deprecated.
2055 If @option{expansion} is set to @code{none}, the text is printed verbatim.
2057 If @option{expansion} is set to @code{normal} (which will be the default),
2058 the following expansion mechanism is used.
2060 The backslash character '\', followed by any character, always expands to
2061 the second character.
2063 Sequence of the form @code{%@{...@}} are expanded. The text between the
2064 braces is a function name, possibly followed by arguments separated by ':'.
2065 If the arguments contain special characters or delimiters (':' or '@}'),
2066 they should be escaped.
2068 Note that they probably must also be escaped as the value for the
2069 @option{text} option in the filter argument string and as the filter
2070 argument in the filter graph description, and possibly also for the shell,
2071 that makes up to four levels of escaping; using a text file avoids these
2074 The following functions are available:
2079 The time at which the filter is running, expressed in UTC.
2080 It can accept an argument: a strftime() format string.
2083 The time at which the filter is running, expressed in the local time zone.
2084 It can accept an argument: a strftime() format string.
2087 The frame number, starting from 0.
2090 The timestamp of the current frame, in seconds, with microsecond accuracy.
2094 @subsection Examples
2096 Some examples follow.
2101 Draw "Test Text" with font FreeSerif, using the default values for the
2102 optional parameters.
2105 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text'"
2109 Draw 'Test Text' with font FreeSerif of size 24 at position x=100
2110 and y=50 (counting from the top-left corner of the screen), text is
2111 yellow with a red box around it. Both the text and the box have an
2115 drawtext="fontfile=/usr/share/fonts/truetype/freefont/FreeSerif.ttf: text='Test Text':\
2116 x=100: y=50: fontsize=24: fontcolor=yellow@@0.2: box=1: boxcolor=red@@0.2"
2119 Note that the double quotes are not necessary if spaces are not used
2120 within the parameter list.
2123 Show the text at the center of the video frame:
2125 drawtext="fontsize=30:fontfile=FreeSerif.ttf:text='hello world':x=(w-text_w)/2:y=(h-text_h-line_h)/2"
2129 Show a text line sliding from right to left in the last row of the video
2130 frame. The file @file{LONG_LINE} is assumed to contain a single line
2133 drawtext="fontsize=15:fontfile=FreeSerif.ttf:text=LONG_LINE:y=h-line_h:x=-50*t"
2137 Show the content of file @file{CREDITS} off the bottom of the frame and scroll up.
2139 drawtext="fontsize=20:fontfile=FreeSerif.ttf:textfile=CREDITS:y=h-20*t"
2143 Draw a single green letter "g", at the center of the input video.
2144 The glyph baseline is placed at half screen height.
2146 drawtext="fontsize=60:fontfile=FreeSerif.ttf:fontcolor=green:text=g:x=(w-max_glyph_w)/2:y=h/2-ascent"
2150 Show text for 1 second every 3 seconds:
2152 drawtext="fontfile=FreeSerif.ttf:fontcolor=white:x=100:y=x/dar:draw=lt(mod(t\,3)\,1):text='blink'"
2156 Use fontconfig to set the font. Note that the colons need to be escaped.
2158 drawtext='fontfile=Linux Libertine O-40\:style=Semibold:text=FFmpeg'
2162 Print the date of a real-time encoding (see strftime(3)):
2164 drawtext='fontfile=FreeSans.ttf:expansion=normal:text=%@{localtime:%a %b %d %Y@}'
2169 For more information about libfreetype, check:
2170 @url{http://www.freetype.org/}.
2172 For more information about fontconfig, check:
2173 @url{http://freedesktop.org/software/fontconfig/fontconfig-user.html}.
2177 Detect and draw edges. The filter uses the Canny Edge Detection algorithm.
2179 This filter accepts the following optional named parameters:
2183 Set low and high threshold values used by the Canny thresholding
2186 The high threshold selects the "strong" edge pixels, which are then
2187 connected through 8-connectivity with the "weak" edge pixels selected
2188 by the low threshold.
2190 @var{low} and @var{high} threshold values must be choosen in the range
2191 [0,1], and @var{low} should be lesser or equal to @var{high}.
2193 Default value for @var{low} is @code{20/255}, and default value for @var{high}
2199 edgedetect=low=0.1:high=0.4
2204 Apply fade-in/out effect to input video.
2206 It accepts the parameters:
2207 @var{type}:@var{start_frame}:@var{nb_frames}[:@var{options}]
2209 @var{type} specifies if the effect type, can be either "in" for
2210 fade-in, or "out" for a fade-out effect.
2212 @var{start_frame} specifies the number of the start frame for starting
2213 to apply the fade effect.
2215 @var{nb_frames} specifies the number of frames for which the fade
2216 effect has to last. At the end of the fade-in effect the output video
2217 will have the same intensity as the input video, at the end of the
2218 fade-out transition the output video will be completely black.
2220 @var{options} is an optional sequence of @var{key}=@var{value} pairs,
2221 separated by ":". The description of the accepted options follows.
2228 @item start_frame, s
2229 See @var{start_frame}.
2232 See @var{nb_frames}.
2235 If set to 1, fade only alpha channel, if one exists on the input.
2239 A few usage examples follow, usable too as test scenarios.
2241 # fade in first 30 frames of video
2244 # fade out last 45 frames of a 200-frame video
2247 # fade in first 25 frames and fade out last 25 frames of a 1000-frame video
2248 fade=in:0:25, fade=out:975:25
2250 # make first 5 frames black, then fade in from frame 5-24
2253 # fade in alpha over first 25 frames of video
2254 fade=in:0:25:alpha=1
2259 Extract a single field from an interlaced image using stride
2260 arithmetic to avoid wasting CPU time. The output frames are marked as
2263 This filter accepts the following named options:
2266 Specify whether to extract the top (if the value is @code{0} or
2267 @code{top}) or the bottom field (if the value is @code{1} or
2271 If the option key is not specified, the first value sets the @var{type}
2272 option. For example:
2284 Transform the field order of the input video.
2286 It accepts one parameter which specifies the required field order that
2287 the input interlaced video will be transformed to. The parameter can
2288 assume one of the following values:
2292 output bottom field first
2294 output top field first
2297 Default value is "tff".
2299 Transformation is achieved by shifting the picture content up or down
2300 by one line, and filling the remaining line with appropriate picture content.
2301 This method is consistent with most broadcast field order converters.
2303 If the input video is not flagged as being interlaced, or it is already
2304 flagged as being of the required output field order then this filter does
2305 not alter the incoming video.
2307 This filter is very useful when converting to or from PAL DV material,
2308 which is bottom field first.
2312 ffmpeg -i in.vob -vf "fieldorder=bff" out.dv
2317 Buffer input images and send them when they are requested.
2319 This filter is mainly useful when auto-inserted by the libavfilter
2322 The filter does not take parameters.
2326 Convert the input video to one of the specified pixel formats.
2327 Libavfilter will try to pick one that is supported for the input to
2330 The filter accepts a list of pixel format names, separated by ":",
2331 for example "yuv420p:monow:rgb24".
2333 Some examples follow:
2335 # convert the input video to the format "yuv420p"
2338 # convert the input video to any of the formats in the list
2339 format=yuv420p:yuv444p:yuv410p
2344 Convert the video to specified constant framerate by duplicating or dropping
2345 frames as necessary.
2347 This filter accepts the following named parameters:
2351 Desired output framerate.
2354 Rounding method. The default is @code{near}.
2360 Select one frame every N.
2362 This filter accepts in input a string representing a positive
2363 integer. Default argument is @code{1}.
2368 Apply a frei0r effect to the input video.
2370 To enable compilation of this filter you need to install the frei0r
2371 header and configure FFmpeg with @code{--enable-frei0r}.
2373 The filter supports the syntax:
2375 @var{filter_name}[@{:|=@}@var{param1}:@var{param2}:...:@var{paramN}]
2378 @var{filter_name} is the name of the frei0r effect to load. If the
2379 environment variable @env{FREI0R_PATH} is defined, the frei0r effect
2380 is searched in each one of the directories specified by the colon (or
2381 semicolon on Windows platforms) separated list in @env{FREIOR_PATH},
2382 otherwise in the standard frei0r paths, which are in this order:
2383 @file{HOME/.frei0r-1/lib/}, @file{/usr/local/lib/frei0r-1/},
2384 @file{/usr/lib/frei0r-1/}.
2386 @var{param1}, @var{param2}, ... , @var{paramN} specify the parameters
2387 for the frei0r effect.
2389 A frei0r effect parameter can be a boolean (whose values are specified
2390 with "y" and "n"), a double, a color (specified by the syntax
2391 @var{R}/@var{G}/@var{B}, @var{R}, @var{G}, and @var{B} being float
2392 numbers from 0.0 to 1.0) or by an @code{av_parse_color()} color
2393 description), a position (specified by the syntax @var{X}/@var{Y},
2394 @var{X} and @var{Y} being float numbers) and a string.
2396 The number and kind of parameters depend on the loaded effect. If an
2397 effect parameter is not specified the default value is set.
2399 Some examples follow:
2403 Apply the distort0r effect, set the first two double parameters:
2405 frei0r=distort0r:0.5:0.01
2409 Apply the colordistance effect, take a color as first parameter:
2411 frei0r=colordistance:0.2/0.3/0.4
2412 frei0r=colordistance:violet
2413 frei0r=colordistance:0x112233
2417 Apply the perspective effect, specify the top left and top right image
2420 frei0r=perspective:0.2/0.2:0.8/0.2
2424 For more information see:
2425 @url{http://frei0r.dyne.org}
2429 The filter takes one, two or three equations as parameter, separated by ':'.
2430 The first equation is mandatory and applies to the luma plane. The two
2431 following are respectively for chroma blue and chroma red planes.
2433 The filter syntax allows named parameters:
2437 the luminance expression
2439 the chrominance blue expression
2441 the chrominance red expression
2444 If one of the chrominance expression is not defined, it falls back on the other
2445 one. If none of them are specified, they will evaluate the luminance
2448 The expressions can use the following variables and functions:
2452 The sequential number of the filtered frame, starting from @code{0}.
2455 The coordinates of the current sample.
2458 The width and height of the image.
2461 Width and height scale depending on the currently filtered plane. It is the
2462 ratio between the corresponding luma plane number of pixels and the current
2463 plane ones. E.g. for YUV4:2:0 the values are @code{1,1} for the luma plane, and
2464 @code{0.5,0.5} for chroma planes.
2467 Return the value of the pixel at location (@var{x},@var{y}) of the current
2471 Return the value of the pixel at location (@var{x},@var{y}) of the luminance
2475 Return the value of the pixel at location (@var{x},@var{y}) of the
2476 blue-difference chroma plane.
2479 Return the value of the pixel at location (@var{x},@var{y}) of the
2480 red-difference chroma plane.
2483 For functions, if @var{x} and @var{y} are outside the area, the value will be
2484 automatically clipped to the closer edge.
2486 Some examples follow:
2490 Flip the image horizontally:
2496 Generate a fancy enigmatic moving light:
2498 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
2504 Fix the banding artifacts that are sometimes introduced into nearly flat
2505 regions by truncation to 8bit color depth.
2506 Interpolate the gradients that should go where the bands are, and
2509 This filter is designed for playback only. Do not use it prior to
2510 lossy compression, because compression tends to lose the dither and
2511 bring back the bands.
2513 The filter takes two optional parameters, separated by ':':
2514 @var{strength}:@var{radius}
2516 @var{strength} is the maximum amount by which the filter will change
2517 any one pixel. Also the threshold for detecting nearly flat
2518 regions. Acceptable values range from .51 to 255, default value is
2519 1.2, out-of-range values will be clipped to the valid range.
2521 @var{radius} is the neighborhood to fit the gradient to. A larger
2522 radius makes for smoother gradients, but also prevents the filter from
2523 modifying the pixels near detailed regions. Acceptable values are
2524 8-32, default value is 16, out-of-range values will be clipped to the
2528 # default parameters
2537 Flip the input video horizontally.
2539 For example to horizontally flip the input video with @command{ffmpeg}:
2541 ffmpeg -i in.avi -vf "hflip" out.avi
2546 High precision/quality 3d denoise filter. This filter aims to reduce
2547 image noise producing smooth images and making still images really
2548 still. It should enhance compressibility.
2550 It accepts the following optional parameters:
2551 @var{luma_spatial}:@var{chroma_spatial}:@var{luma_tmp}:@var{chroma_tmp}
2555 a non-negative float number which specifies spatial luma strength,
2558 @item chroma_spatial
2559 a non-negative float number which specifies spatial chroma strength,
2560 defaults to 3.0*@var{luma_spatial}/4.0
2563 a float number which specifies luma temporal strength, defaults to
2564 6.0*@var{luma_spatial}/4.0
2567 a float number which specifies chroma temporal strength, defaults to
2568 @var{luma_tmp}*@var{chroma_spatial}/@var{luma_spatial}
2573 Modify the hue and/or the saturation of the input.
2575 This filter accepts the following optional named options:
2579 Specify the hue angle as a number of degrees. It accepts a float
2580 number or an expression, and defaults to 0.0.
2583 Specify the hue angle as a number of degrees. It accepts a float
2584 number or an expression, and defaults to 0.0.
2587 Specify the saturation in the [-10,10] range. It accepts a float number and
2591 The @var{h}, @var{H} and @var{s} parameters are expressions containing the
2592 following constants:
2596 frame count of the input frame starting from 0
2599 presentation timestamp of the input frame expressed in time base units
2602 frame rate of the input video, NAN if the input frame rate is unknown
2605 timestamp expressed in seconds, NAN if the input timestamp is unknown
2608 time base of the input video
2611 The options can also be set using the syntax: @var{hue}:@var{saturation}
2613 In this case @var{hue} is expressed in degrees.
2615 Some examples follow:
2618 Set the hue to 90 degrees and the saturation to 1.0:
2624 Same command but expressing the hue in radians:
2630 Same command without named options, hue must be expressed in degrees:
2636 Note that "h:s" syntax does not support expressions for the values of
2637 h and s, so the following example will issue an error:
2643 Rotate hue and make the saturation swing between 0
2644 and 2 over a period of 1 second:
2646 hue="H=2*PI*t: s=sin(2*PI*t)+1"
2650 Apply a 3 seconds saturation fade-in effect starting at 0:
2655 The general fade-in expression can be written as:
2657 hue="s=min(0\, max((t-START)/DURATION\, 1))"
2661 Apply a 3 seconds saturation fade-out effect starting at 5 seconds:
2663 hue="s=max(0\, min(1\, (8-t)/3))"
2666 The general fade-out expression can be written as:
2668 hue="s=max(0\, min(1\, (START+DURATION-t)/DURATION))"
2673 @subsection Commands
2675 This filter supports the following command:
2678 Modify the hue and/or the saturation of the input video.
2679 The command accepts the same named options and syntax than when calling the
2680 filter from the command-line.
2682 If a parameter is omitted, it is kept at its current value.
2687 Interlaceing detect filter. This filter tries to detect if the input is
2688 interlaced or progressive. Top or bottom field first.
2690 @section lut, lutrgb, lutyuv
2692 Compute a look-up table for binding each pixel component input value
2693 to an output value, and apply it to input video.
2695 @var{lutyuv} applies a lookup table to a YUV input video, @var{lutrgb}
2696 to an RGB input video.
2698 These filters accept in input a ":"-separated list of options, which
2699 specify the expressions used for computing the lookup table for the
2700 corresponding pixel component values.
2702 The @var{lut} filter requires either YUV or RGB pixel formats in
2703 input, and accepts the options:
2705 @item @var{c0} (first pixel component)
2706 @item @var{c1} (second pixel component)
2707 @item @var{c2} (third pixel component)
2708 @item @var{c3} (fourth pixel component, corresponds to the alpha component)
2711 The exact component associated to each option depends on the format in
2714 The @var{lutrgb} filter requires RGB pixel formats in input, and
2715 accepts the options:
2717 @item @var{r} (red component)
2718 @item @var{g} (green component)
2719 @item @var{b} (blue component)
2720 @item @var{a} (alpha component)
2723 The @var{lutyuv} filter requires YUV pixel formats in input, and
2724 accepts the options:
2726 @item @var{y} (Y/luminance component)
2727 @item @var{u} (U/Cb component)
2728 @item @var{v} (V/Cr component)
2729 @item @var{a} (alpha component)
2732 The expressions can contain the following constants and functions:
2736 the input width and height
2739 input value for the pixel component
2742 the input value clipped in the @var{minval}-@var{maxval} range
2745 maximum value for the pixel component
2748 minimum value for the pixel component
2751 the negated value for the pixel component value clipped in the
2752 @var{minval}-@var{maxval} range , it corresponds to the expression
2753 "maxval-clipval+minval"
2756 the computed value in @var{val} clipped in the
2757 @var{minval}-@var{maxval} range
2759 @item gammaval(gamma)
2760 the computed gamma correction value of the pixel component value
2761 clipped in the @var{minval}-@var{maxval} range, corresponds to the
2763 "pow((clipval-minval)/(maxval-minval)\,@var{gamma})*(maxval-minval)+minval"
2767 All expressions default to "val".
2769 Some examples follow:
2771 # negate input video
2772 lutrgb="r=maxval+minval-val:g=maxval+minval-val:b=maxval+minval-val"
2773 lutyuv="y=maxval+minval-val:u=maxval+minval-val:v=maxval+minval-val"
2775 # the above is the same as
2776 lutrgb="r=negval:g=negval:b=negval"
2777 lutyuv="y=negval:u=negval:v=negval"
2782 # remove chroma components, turns the video into a graytone image
2783 lutyuv="u=128:v=128"
2785 # apply a luma burning effect
2788 # remove green and blue components
2791 # set a constant alpha channel value on input
2792 format=rgba,lutrgb=a="maxval-minval/2"
2794 # correct luminance gamma by a 0.5 factor
2795 lutyuv=y=gammaval(0.5)
2800 Apply an MPlayer filter to the input video.
2802 This filter provides a wrapper around most of the filters of
2805 This wrapper is considered experimental. Some of the wrapped filters
2806 may not work properly and we may drop support for them, as they will
2807 be implemented natively into FFmpeg. Thus you should avoid
2808 depending on them when writing portable scripts.
2810 The filters accepts the parameters:
2811 @var{filter_name}[:=]@var{filter_params}
2813 @var{filter_name} is the name of a supported MPlayer filter,
2814 @var{filter_params} is a string containing the parameters accepted by
2817 The list of the currently supported filters follows:
2852 The parameter syntax and behavior for the listed filters are the same
2853 of the corresponding MPlayer filters. For detailed instructions check
2854 the "VIDEO FILTERS" section in the MPlayer manual.
2856 Some examples follow:
2859 Adjust gamma, brightness, contrast:
2865 Add temporal noise to input video:
2871 See also mplayer(1), @url{http://www.mplayerhq.hu/}.
2877 This filter accepts an integer in input, if non-zero it negates the
2878 alpha component (if available). The default value in input is 0.
2882 Force libavfilter not to use any of the specified pixel formats for the
2883 input to the next filter.
2885 The filter accepts a list of pixel format names, separated by ":",
2886 for example "yuv420p:monow:rgb24".
2888 Some examples follow:
2890 # force libavfilter to use a format different from "yuv420p" for the
2891 # input to the vflip filter
2892 noformat=yuv420p,vflip
2894 # convert the input video to any of the formats not contained in the list
2895 noformat=yuv420p:yuv444p:yuv410p
2900 Pass the video source unchanged to the output.
2904 Apply video transform using libopencv.
2906 To enable this filter install libopencv library and headers and
2907 configure FFmpeg with @code{--enable-libopencv}.
2909 The filter takes the parameters: @var{filter_name}@{:=@}@var{filter_params}.
2911 @var{filter_name} is the name of the libopencv filter to apply.
2913 @var{filter_params} specifies the parameters to pass to the libopencv
2914 filter. If not specified the default values are assumed.
2916 Refer to the official libopencv documentation for more precise
2918 @url{http://opencv.willowgarage.com/documentation/c/image_filtering.html}
2920 Follows the list of supported libopencv filters.
2925 Dilate an image by using a specific structuring element.
2926 This filter corresponds to the libopencv function @code{cvDilate}.
2928 It accepts the parameters: @var{struct_el}:@var{nb_iterations}.
2930 @var{struct_el} represents a structuring element, and has the syntax:
2931 @var{cols}x@var{rows}+@var{anchor_x}x@var{anchor_y}/@var{shape}
2933 @var{cols} and @var{rows} represent the number of columns and rows of
2934 the structuring element, @var{anchor_x} and @var{anchor_y} the anchor
2935 point, and @var{shape} the shape for the structuring element, and
2936 can be one of the values "rect", "cross", "ellipse", "custom".
2938 If the value for @var{shape} is "custom", it must be followed by a
2939 string of the form "=@var{filename}". The file with name
2940 @var{filename} is assumed to represent a binary image, with each
2941 printable character corresponding to a bright pixel. When a custom
2942 @var{shape} is used, @var{cols} and @var{rows} are ignored, the number
2943 or columns and rows of the read file are assumed instead.
2945 The default value for @var{struct_el} is "3x3+0x0/rect".
2947 @var{nb_iterations} specifies the number of times the transform is
2948 applied to the image, and defaults to 1.
2950 Follow some example:
2952 # use the default values
2955 # dilate using a structuring element with a 5x5 cross, iterate two times
2956 ocv=dilate=5x5+2x2/cross:2
2958 # read the shape from the file diamond.shape, iterate two times
2959 # the file diamond.shape may contain a pattern of characters like this:
2965 # the specified cols and rows are ignored (but not the anchor point coordinates)
2966 ocv=0x0+2x2/custom=diamond.shape:2
2971 Erode an image by using a specific structuring element.
2972 This filter corresponds to the libopencv function @code{cvErode}.
2974 The filter accepts the parameters: @var{struct_el}:@var{nb_iterations},
2975 with the same syntax and semantics as the @ref{dilate} filter.
2979 Smooth the input video.
2981 The filter takes the following parameters:
2982 @var{type}:@var{param1}:@var{param2}:@var{param3}:@var{param4}.
2984 @var{type} is the type of smooth filter to apply, and can be one of
2985 the following values: "blur", "blur_no_scale", "median", "gaussian",
2986 "bilateral". The default value is "gaussian".
2988 @var{param1}, @var{param2}, @var{param3}, and @var{param4} are
2989 parameters whose meanings depend on smooth type. @var{param1} and
2990 @var{param2} accept integer positive values or 0, @var{param3} and
2991 @var{param4} accept float values.
2993 The default value for @var{param1} is 3, the default value for the
2994 other parameters is 0.
2996 These parameters correspond to the parameters assigned to the
2997 libopencv function @code{cvSmooth}.
3002 Overlay one video on top of another.
3004 It takes two inputs and one output, the first input is the "main"
3005 video on which the second input is overlayed.
3007 It accepts the parameters: @var{x}:@var{y}[:@var{options}].
3009 @var{x} is the x coordinate of the overlayed video on the main video,
3010 @var{y} is the y coordinate. @var{x} and @var{y} are expressions containing
3011 the following parameters:
3014 @item main_w, main_h
3015 main input width and height
3018 same as @var{main_w} and @var{main_h}
3020 @item overlay_w, overlay_h
3021 overlay input width and height
3024 same as @var{overlay_w} and @var{overlay_h}
3027 @var{options} is an optional list of @var{key}=@var{value} pairs,
3030 The description of the accepted options follows.
3034 If set to 1, force the filter to accept inputs in the RGB
3035 color space. Default value is 0.
3038 Be aware that frames are taken from each input video in timestamp
3039 order, hence, if their initial timestamps differ, it is a a good idea
3040 to pass the two inputs through a @var{setpts=PTS-STARTPTS} filter to
3041 have them begin in the same zero timestamp, as it does the example for
3042 the @var{movie} filter.
3044 Follow some examples:
3046 # draw the overlay at 10 pixels from the bottom right
3047 # corner of the main video.
3048 overlay=main_w-overlay_w-10:main_h-overlay_h-10
3050 # insert a transparent PNG logo in the bottom left corner of the input
3051 ffmpeg -i input -i logo -filter_complex 'overlay=10:main_h-overlay_h-10' output
3053 # insert 2 different transparent PNG logos (second logo on bottom
3055 ffmpeg -i input -i logo1 -i logo2 -filter_complex
3056 'overlay=10:H-h-10,overlay=W-w-10:H-h-10' output
3058 # add a transparent color layer on top of the main video,
3059 # WxH specifies the size of the main input to the overlay filter
3060 color=red@@.3:WxH [over]; [in][over] overlay [out]
3062 # play an original video and a filtered version (here with the deshake filter)
3064 ffplay input.avi -vf 'split[a][b]; [a]pad=iw*2:ih[src]; [b]deshake[filt]; [src][filt]overlay=w'
3066 # the previous example is the same as:
3067 ffplay input.avi -vf 'split[b], pad=iw*2[src], [b]deshake, [src]overlay=w'
3070 You can chain together more overlays but the efficiency of such
3071 approach is yet to be tested.
3075 Add paddings to the input image, and places the original input at the
3076 given coordinates @var{x}, @var{y}.
3078 It accepts the following parameters:
3079 @var{width}:@var{height}:@var{x}:@var{y}:@var{color}.
3081 The parameters @var{width}, @var{height}, @var{x}, and @var{y} are
3082 expressions containing the following constants:
3086 the input video width and height
3089 same as @var{in_w} and @var{in_h}
3092 the output width and height, that is the size of the padded area as
3093 specified by the @var{width} and @var{height} expressions
3096 same as @var{out_w} and @var{out_h}
3099 x and y offsets as specified by the @var{x} and @var{y}
3100 expressions, or NAN if not yet specified
3103 same as @var{iw} / @var{ih}
3106 input sample aspect ratio
3109 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
3112 horizontal and vertical chroma subsample values. For example for the
3113 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3116 Follows the description of the accepted parameters.
3121 Specify the size of the output image with the paddings added. If the
3122 value for @var{width} or @var{height} is 0, the corresponding input size
3123 is used for the output.
3125 The @var{width} expression can reference the value set by the
3126 @var{height} expression, and vice versa.
3128 The default value of @var{width} and @var{height} is 0.
3132 Specify the offsets where to place the input image in the padded area
3133 with respect to the top/left border of the output image.
3135 The @var{x} expression can reference the value set by the @var{y}
3136 expression, and vice versa.
3138 The default value of @var{x} and @var{y} is 0.
3142 Specify the color of the padded area, it can be the name of a color
3143 (case insensitive match) or a 0xRRGGBB[AA] sequence.
3145 The default value of @var{color} is "black".
3149 @subsection Examples
3153 Add paddings with color "violet" to the input video. Output video
3154 size is 640x480, the top-left corner of the input video is placed at
3157 pad=640:480:0:40:violet
3161 Pad the input to get an output with dimensions increased by 3/2,
3162 and put the input video at the center of the padded area:
3164 pad="3/2*iw:3/2*ih:(ow-iw)/2:(oh-ih)/2"
3168 Pad the input to get a squared output with size equal to the maximum
3169 value between the input width and height, and put the input video at
3170 the center of the padded area:
3172 pad="max(iw\,ih):ow:(ow-iw)/2:(oh-ih)/2"
3176 Pad the input to get a final w/h ratio of 16:9:
3178 pad="ih*16/9:ih:(ow-iw)/2:(oh-ih)/2"
3182 In case of anamorphic video, in order to set the output display aspect
3183 correctly, it is necessary to use @var{sar} in the expression,
3184 according to the relation:
3186 (ih * X / ih) * sar = output_dar
3187 X = output_dar / sar
3190 Thus the previous example needs to be modified to:
3192 pad="ih*16/9/sar:ih:(ow-iw)/2:(oh-ih)/2"
3196 Double output size and put the input video in the bottom-right
3197 corner of the output padded area:
3199 pad="2*iw:2*ih:ow-iw:oh-ih"
3203 @section pixdesctest
3205 Pixel format descriptor test filter, mainly useful for internal
3206 testing. The output video should be equal to the input video.
3210 format=monow, pixdesctest
3213 can be used to test the monowhite pixel format descriptor definition.
3217 Suppress a TV station logo, using an image file to determine which
3218 pixels comprise the logo. It works by filling in the pixels that
3219 comprise the logo with neighboring pixels.
3221 This filter requires one argument which specifies the filter bitmap
3222 file, which can be any image format supported by libavformat. The
3223 width and height of the image file must match those of the video
3224 stream being processed.
3226 Pixels in the provided bitmap image with a value of zero are not
3227 considered part of the logo, non-zero pixels are considered part of
3228 the logo. If you use white (255) for the logo and black (0) for the
3229 rest, you will be safe. For making the filter bitmap, it is
3230 recommended to take a screen capture of a black frame with the logo
3231 visible, and then using a threshold filter followed by the erode
3232 filter once or twice.
3234 If needed, little splotches can be fixed manually. Remember that if
3235 logo pixels are not covered, the filter quality will be much
3236 reduced. Marking too many pixels as part of the logo does not hurt as
3237 much, but it will increase the amount of blurring needed to cover over
3238 the image and will destroy more information than necessary, and extra
3239 pixels will slow things down on a large logo.
3243 Scale (resize) the input video, using the libswscale library.
3245 The scale filter forces the output display aspect ratio to be the same
3246 of the input, by changing the output sample aspect ratio.
3248 This filter accepts a list of named options in the form of
3249 @var{key}=@var{value} pairs separated by ":". If the key for the first
3250 two options is not specified, the assumed keys for the first two
3251 values are @code{w} and @code{h}. If the first option has no key and
3252 can be interpreted like a video size specification, it will be used
3253 to set the video size.
3255 A description of the accepted options follows.
3259 Set the video width expression, default value is @code{iw}. See below
3260 for the list of accepted constants.
3263 Set the video heiht expression, default value is @code{ih}.
3264 See below for the list of accepted constants.
3267 Set the interlacing. It accepts the following values:
3271 force interlaced aware scaling
3274 do not apply interlaced scaling
3277 select interlaced aware scaling depending on whether the source frames
3278 are flagged as interlaced or not
3281 Default value is @code{0}.
3284 Set libswscale scaling flags. If not explictly specified the filter
3285 applies a bilinear scaling algorithm.
3288 Set the video size, the value must be a valid abbreviation or in the
3289 form @var{width}x@var{height}.
3292 The values of the @var{w} and @var{h} options are expressions
3293 containing the following constants:
3297 the input width and height
3300 same as @var{in_w} and @var{in_h}
3303 the output (cropped) width and height
3306 same as @var{out_w} and @var{out_h}
3309 same as @var{iw} / @var{ih}
3312 input sample aspect ratio
3315 input display aspect ratio, it is the same as (@var{iw} / @var{ih}) * @var{sar}
3318 horizontal and vertical chroma subsample values. For example for the
3319 pixel format "yuv422p" @var{hsub} is 2 and @var{vsub} is 1.
3322 If the input image format is different from the format requested by
3323 the next filter, the scale filter will convert the input to the
3326 If the value for @var{width} or @var{height} is 0, the respective input
3327 size is used for the output.
3329 If the value for @var{width} or @var{height} is -1, the scale filter will
3330 use, for the respective output size, a value that maintains the aspect
3331 ratio of the input image.
3333 @subsection Examples
3337 Scale the input video to a size of 200x100:
3342 This is equivalent to:
3353 Specify a size abbreviation for the output size:
3358 which can also be written as:
3364 Scale the input to 2x:
3370 The above is the same as:
3376 Scale the input to 2x with forced interlaced scaling:
3378 scale=2*iw:2*ih:interl=1
3382 Scale the input to half size:
3388 Increase the width, and set the height to the same size:
3394 Seek for Greek harmony:
3401 Increase the height, and set the width to 3/2 of the height:
3407 Increase the size, but make the size a multiple of the chroma:
3409 scale="trunc(3/2*iw/hsub)*hsub:trunc(3/2*ih/vsub)*vsub"
3413 Increase the width to a maximum of 500 pixels, keep the same input
3416 scale='min(500\, iw*3/2):-1'
3421 Select frames to pass in output.
3423 It accepts in input an expression, which is evaluated for each input
3424 frame. If the expression is evaluated to a non-zero value, the frame
3425 is selected and passed to the output, otherwise it is discarded.
3427 The expression can contain the following constants:
3431 the sequential number of the filtered frame, starting from 0
3434 the sequential number of the selected frame, starting from 0
3436 @item prev_selected_n
3437 the sequential number of the last selected frame, NAN if undefined
3440 timebase of the input timestamps
3443 the PTS (Presentation TimeStamp) of the filtered video frame,
3444 expressed in @var{TB} units, NAN if undefined
3447 the PTS (Presentation TimeStamp) of the filtered video frame,
3448 expressed in seconds, NAN if undefined
3451 the PTS of the previously filtered video frame, NAN if undefined
3453 @item prev_selected_pts
3454 the PTS of the last previously filtered video frame, NAN if undefined
3456 @item prev_selected_t
3457 the PTS of the last previously selected video frame, NAN if undefined
3460 the PTS of the first video frame in the video, NAN if undefined
3463 the time of the first video frame in the video, NAN if undefined
3466 the type of the filtered frame, can assume one of the following
3478 @item interlace_type
3479 the frame interlace type, can assume one of the following values:
3482 the frame is progressive (not interlaced)
3484 the frame is top-field-first
3486 the frame is bottom-field-first
3490 1 if the filtered frame is a key-frame, 0 otherwise
3493 the position in the file of the filtered frame, -1 if the information
3494 is not available (e.g. for synthetic video)
3497 value between 0 and 1 to indicate a new scene; a low value reflects a low
3498 probability for the current frame to introduce a new scene, while a higher
3499 value means the current frame is more likely to be one (see the example below)
3503 The default value of the select expression is "1".
3505 Some examples follow:
3508 # select all frames in input
3511 # the above is the same as:
3517 # select only I-frames
3518 select='eq(pict_type\,I)'
3520 # select one frame every 100
3521 select='not(mod(n\,100))'
3523 # select only frames contained in the 10-20 time interval
3524 select='gte(t\,10)*lte(t\,20)'
3526 # select only I frames contained in the 10-20 time interval
3527 select='gte(t\,10)*lte(t\,20)*eq(pict_type\,I)'
3529 # select frames with a minimum distance of 10 seconds
3530 select='isnan(prev_selected_t)+gte(t-prev_selected_t\,10)'
3533 Complete example to create a mosaic of the first scenes:
3536 ffmpeg -i video.avi -vf select='gt(scene\,0.4)',scale=160:120,tile -frames:v 1 preview.png
3539 Comparing @var{scene} against a value between 0.3 and 0.5 is generally a sane
3542 @section setdar, setsar
3544 The @code{setdar} filter sets the Display Aspect Ratio for the filter
3547 This is done by changing the specified Sample (aka Pixel) Aspect
3548 Ratio, according to the following equation:
3550 @var{DAR} = @var{HORIZONTAL_RESOLUTION} / @var{VERTICAL_RESOLUTION} * @var{SAR}
3553 Keep in mind that the @code{setdar} filter does not modify the pixel
3554 dimensions of the video frame. Also the display aspect ratio set by
3555 this filter may be changed by later filters in the filterchain,
3556 e.g. in case of scaling or if another "setdar" or a "setsar" filter is
3559 The @code{setsar} filter sets the Sample (aka Pixel) Aspect Ratio for
3560 the filter output video.
3562 Note that as a consequence of the application of this filter, the
3563 output display aspect ratio will change according to the equation
3566 Keep in mind that the sample aspect ratio set by the @code{setsar}
3567 filter may be changed by later filters in the filterchain, e.g. if
3568 another "setsar" or a "setdar" filter is applied.
3570 The @code{setdar} and @code{setsar} filters accept a string in the
3571 form @var{num}:@var{den} expressing an aspect ratio, or the following
3572 named options, expressed as a sequence of @var{key}=@var{value} pairs,
3577 Set the maximum integer value to use for expressing numerator and
3578 denominator when reducing the expressed aspect ratio to a rational.
3579 Default value is @code{100}.
3582 Set the aspect ratio used by the filter.
3584 The parameter can be a floating point number string, an expression, or
3585 a string of the form @var{num}:@var{den}, where @var{num} and
3586 @var{den} are the numerator and denominator of the aspect ratio. If
3587 the parameter is not specified, it is assumed the value "0".
3588 In case the form "@var{num}:@var{den}" the @code{:} character should
3592 If the keys are omitted in the named options list, the specifed values
3593 are assumed to be @var{ratio} and @var{max} in that order.
3595 For example to change the display aspect ratio to 16:9, specify:
3600 The example above is equivalent to:
3605 To change the sample aspect ratio to 10:11, specify:
3610 To set a display aspect ratio of 16:9, and specify a maximum integer value of
3611 1000 in the aspect ratio reduction, use the command:
3613 setdar=ratio='16:9':max=1000
3618 Force field for the output video frame.
3620 The @code{setfield} filter marks the interlace type field for the
3621 output frames. It does not change the input frame, but only sets the
3622 corresponding property, which affects how the frame is treated by
3623 following filters (e.g. @code{fieldorder} or @code{yadif}).
3625 It accepts a string parameter, which can assume the following values:
3628 Keep the same field property.
3631 Mark the frame as bottom-field-first.
3634 Mark the frame as top-field-first.
3637 Mark the frame as progressive.
3642 Show a line containing various information for each input video frame.
3643 The input video is not modified.
3645 The shown line contains a sequence of key/value pairs of the form
3646 @var{key}:@var{value}.
3648 A description of each shown parameter follows:
3652 sequential number of the input frame, starting from 0
3655 Presentation TimeStamp of the input frame, expressed as a number of
3656 time base units. The time base unit depends on the filter input pad.
3659 Presentation TimeStamp of the input frame, expressed as a number of
3663 position of the frame in the input stream, -1 if this information in
3664 unavailable and/or meaningless (for example in case of synthetic video)
3670 sample aspect ratio of the input frame, expressed in the form
3674 size of the input frame, expressed in the form
3675 @var{width}x@var{height}
3678 interlaced mode ("P" for "progressive", "T" for top field first, "B"
3679 for bottom field first)
3682 1 if the frame is a key frame, 0 otherwise
3685 picture type of the input frame ("I" for an I-frame, "P" for a
3686 P-frame, "B" for a B-frame, "?" for unknown type).
3687 Check also the documentation of the @code{AVPictureType} enum and of
3688 the @code{av_get_picture_type_char} function defined in
3689 @file{libavutil/avutil.h}.
3692 Adler-32 checksum (printed in hexadecimal) of all the planes of the input frame
3694 @item plane_checksum
3695 Adler-32 checksum (printed in hexadecimal) of each plane of the input frame,
3696 expressed in the form "[@var{c0} @var{c1} @var{c2} @var{c3}]"
3701 Pass the images of input video on to next video filter as multiple
3705 ffmpeg -i in.avi -vf "slicify=32" out.avi
3708 The filter accepts the slice height as parameter. If the parameter is
3709 not specified it will use the default value of 16.
3711 Adding this in the beginning of filter chains should make filtering
3712 faster due to better use of the memory cache.
3716 Blur the input video without impacting the outlines.
3718 The filter accepts the following parameters:
3719 @var{luma_radius}:@var{luma_strength}:@var{luma_threshold}[:@var{chroma_radius}:@var{chroma_strength}:@var{chroma_threshold}]
3721 Parameters prefixed by @var{luma} indicate that they work on the
3722 luminance of the pixels whereas parameters prefixed by @var{chroma}
3723 refer to the chrominance of the pixels.
3725 If the chroma parameters are not set, the luma parameters are used for
3726 either the luminance and the chrominance of the pixels.
3728 @var{luma_radius} or @var{chroma_radius} must be a float number in the
3729 range [0.1,5.0] that specifies the variance of the gaussian filter
3730 used to blur the image (slower if larger).
3732 @var{luma_strength} or @var{chroma_strength} must be a float number in
3733 the range [-1.0,1.0] that configures the blurring. A value included in
3734 [0.0,1.0] will blur the image whereas a value included in [-1.0,0.0]
3735 will sharpen the image.
3737 @var{luma_threshold} or @var{chroma_threshold} must be an integer in
3738 the range [-30,30] that is used as a coefficient to determine whether
3739 a pixel should be blurred or not. A value of 0 will filter all the
3740 image, a value included in [0,30] will filter flat areas and a value
3741 included in [-30,0] will filter edges.
3745 Split input video into several identical outputs.
3747 The filter accepts a single parameter which specifies the number of outputs. If
3748 unspecified, it defaults to 2.
3752 ffmpeg -i INPUT -filter_complex split=5 OUTPUT
3754 will create 5 copies of the input video.
3758 [in] split [splitout1][splitout2];
3759 [splitout1] crop=100:100:0:0 [cropout];
3760 [splitout2] pad=200:200:100:100 [padout];
3763 will create two separate outputs from the same input, one cropped and
3768 Scale the input by 2x and smooth using the Super2xSaI (Scale and
3769 Interpolate) pixel art scaling algorithm.
3771 Useful for enlarging pixel art images without reducing sharpness.
3777 Select the most representative frame in a given sequence of consecutive frames.
3779 It accepts as argument the frames batch size to analyze (default @var{N}=100);
3780 in a set of @var{N} frames, the filter will pick one of them, and then handle
3781 the next batch of @var{N} frames until the end.
3783 Since the filter keeps track of the whole frames sequence, a bigger @var{N}
3784 value will result in a higher memory usage, so a high value is not recommended.
3786 The following example extract one picture each 50 frames:
3791 Complete example of a thumbnail creation with @command{ffmpeg}:
3793 ffmpeg -i in.avi -vf thumbnail,scale=300:200 -frames:v 1 out.png
3798 Tile several successive frames together.
3800 It accepts a list of options in the form of @var{key}=@var{value} pairs
3801 separated by ":". A description of the accepted options follows.
3806 Set the grid size (i.e. the number of lines and columns) in the form
3810 Set the outer border margin in pixels.
3813 Set the inner border thickness (i.e. the number of pixels between frames). For
3814 more advanced padding options (such as having different values for the edges),
3815 refer to the pad video filter.
3818 Set the maximum number of frames to render in the given area. It must be less
3819 than or equal to @var{w}x@var{h}. The default value is @code{0}, meaning all
3820 the area will be used.
3824 Alternatively, the options can be specified as a flat string:
3826 @var{layout}[:@var{nb_frames}[:@var{margin}[:@var{padding}]]]
3828 For example, produce 8×8 PNG tiles of all keyframes (@option{-skip_frame
3831 ffmpeg -skip_frame nokey -i file.avi -vf 'scale=128:72,tile=8x8' -an -vsync 0 keyframes%03d.png
3833 The @option{-vsync 0} is necessary to prevent @command{ffmpeg} from
3834 duplicating each output frame to accomodate the originally detected frame
3837 Another example to display @code{5} pictures in an area of @code{3x2} frames,
3838 with @code{7} pixels between them, and @code{2} pixels of initial margin, using
3839 mixed flat and named options:
3841 tile=3x2:nb_frames=5:padding=7:margin=2
3846 Perform various types of temporal field interlacing.
3848 Frames are counted starting from 1, so the first input frame is
3851 This filter accepts a single parameter specifying the mode. Available
3856 Move odd frames into the upper field, even into the lower field,
3857 generating a double height frame at half framerate.
3860 Only output even frames, odd frames are dropped, generating a frame with
3861 unchanged height at half framerate.
3864 Only output odd frames, even frames are dropped, generating a frame with
3865 unchanged height at half framerate.
3868 Expand each frame to full height, but pad alternate lines with black,
3869 generating a frame with double height at the same input framerate.
3871 @item interleave_top, 4
3872 Interleave the upper field from odd frames with the lower field from
3873 even frames, generating a frame with unchanged height at half framerate.
3875 @item interleave_bottom, 5
3876 Interleave the lower field from odd frames with the upper field from
3877 even frames, generating a frame with unchanged height at half framerate.
3879 @item interlacex2, 6
3880 Double frame rate with unchanged height. Frames are inserted each
3881 containing the second temporal field from the previous input frame and
3882 the first temporal field from the next input frame. This mode relies on
3883 the top_field_first flag. Useful for interlaced video displays with no
3884 field synchronisation.
3887 Numeric values are deprecated but are accepted for backward
3888 compatibility reasons.
3890 Default mode is @code{merge}.
3894 Transpose rows with columns in the input video and optionally flip it.
3896 This filter accepts the following named parameters:
3900 Specify the transposition direction. Can assume the following values:
3904 Rotate by 90 degrees counterclockwise and vertically flip (default), that is:
3912 Rotate by 90 degrees clockwise, that is:
3920 Rotate by 90 degrees counterclockwise, that is:
3928 Rotate by 90 degrees clockwise and vertically flip, that is:
3936 For values between 4-7, the transposition is only done if the input
3937 video geometry is portrait and not landscape. These values are
3938 deprecated, the @code{passthrough} option should be used instead.
3941 Do not apply the transposition if the input geometry matches the one
3942 specified by the specified value. It accepts the following values:
3945 Always apply transposition.
3947 Preserve portrait geometry (when @var{height} >= @var{width}).
3949 Preserve landscape geometry (when @var{width} >= @var{height}).
3952 Default value is @code{none}.
3957 Sharpen or blur the input video.
3959 It accepts the following parameters:
3960 @var{luma_msize_x}:@var{luma_msize_y}:@var{luma_amount}:@var{chroma_msize_x}:@var{chroma_msize_y}:@var{chroma_amount}
3962 Negative values for the amount will blur the input video, while positive
3963 values will sharpen. All parameters are optional and default to the
3964 equivalent of the string '5:5:1.0:5:5:0.0'.
3969 Set the luma matrix horizontal size. It can be an integer between 3
3970 and 13, default value is 5.
3973 Set the luma matrix vertical size. It can be an integer between 3
3974 and 13, default value is 5.
3977 Set the luma effect strength. It can be a float number between -2.0
3978 and 5.0, default value is 1.0.
3980 @item chroma_msize_x
3981 Set the chroma matrix horizontal size. It can be an integer between 3
3982 and 13, default value is 5.
3984 @item chroma_msize_y
3985 Set the chroma matrix vertical size. It can be an integer between 3
3986 and 13, default value is 5.
3989 Set the chroma effect strength. It can be a float number between -2.0
3990 and 5.0, default value is 0.0.
3995 # Strong luma sharpen effect parameters
3998 # Strong blur of both luma and chroma parameters
3999 unsharp=7:7:-2:7:7:-2
4001 # Use the default values with @command{ffmpeg}
4002 ffmpeg -i in.avi -vf "unsharp" out.mp4
4007 Flip the input video vertically.
4010 ffmpeg -i in.avi -vf "vflip" out.avi
4015 Deinterlace the input video ("yadif" means "yet another deinterlacing
4018 It accepts the optional parameters: @var{mode}:@var{parity}:@var{auto}.
4020 @var{mode} specifies the interlacing mode to adopt, accepts one of the
4025 output 1 frame for each frame
4027 output 1 frame for each field
4029 like 0 but skips spatial interlacing check
4031 like 1 but skips spatial interlacing check
4036 @var{parity} specifies the picture field parity assumed for the input
4037 interlaced video, accepts one of the following values:
4041 assume top field first
4043 assume bottom field first
4045 enable automatic detection
4048 Default value is -1.
4049 If interlacing is unknown or decoder does not export this information,
4050 top field first will be assumed.
4052 @var{auto} specifies if deinterlacer should trust the interlaced flag
4053 and only deinterlace frames marked as interlaced
4057 deinterlace all frames
4059 only deinterlace frames marked as interlaced
4064 @c man end VIDEO FILTERS
4066 @chapter Video Sources
4067 @c man begin VIDEO SOURCES
4069 Below is a description of the currently available video sources.
4073 Buffer video frames, and make them available to the filter chain.
4075 This source is mainly intended for a programmatic use, in particular
4076 through the interface defined in @file{libavfilter/vsrc_buffer.h}.
4078 It accepts a list of options in the form of @var{key}=@var{value} pairs
4079 separated by ":". A description of the accepted options follows.
4084 Specify the size (width and height) of the buffered video frames.
4087 A string representing the pixel format of the buffered video frames.
4088 It may be a number corresponding to a pixel format, or a pixel format
4092 Specify the timebase assumed by the timestamps of the buffered frames.
4095 Specify the frame rate expected for the video stream.
4098 Specify the sample aspect ratio assumed by the video frames.
4101 Specify the optional parameters to be used for the scale filter which
4102 is automatically inserted when an input change is detected in the
4103 input size or format.
4108 buffer=size=320x240:pix_fmt=yuv410p:time_base=1/24:pixel_aspect=1/1
4111 will instruct the source to accept video frames with size 320x240 and
4112 with format "yuv410p", assuming 1/24 as the timestamps timebase and
4113 square pixels (1:1 sample aspect ratio).
4114 Since the pixel format with name "yuv410p" corresponds to the number 6
4115 (check the enum AVPixelFormat definition in @file{libavutil/pixfmt.h}),
4116 this example corresponds to:
4118 buffer=size=320x240:pixfmt=6:time_base=1/24:pixel_aspect=1/1
4121 Alternatively, the options can be specified as a flat string, but this
4122 syntax is deprecated:
4124 @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}]
4128 Create a pattern generated by an elementary cellular automaton.
4130 The initial state of the cellular automaton can be defined through the
4131 @option{filename}, and @option{pattern} options. If such options are
4132 not specified an initial state is created randomly.
4134 At each new frame a new row in the video is filled with the result of
4135 the cellular automaton next generation. The behavior when the whole
4136 frame is filled is defined by the @option{scroll} option.
4138 This source accepts a list of options in the form of
4139 @var{key}=@var{value} pairs separated by ":". A description of the
4140 accepted options follows.
4144 Read the initial cellular automaton state, i.e. the starting row, from
4146 In the file, each non-whitespace character is considered an alive
4147 cell, a newline will terminate the row, and further characters in the
4148 file will be ignored.
4151 Read the initial cellular automaton state, i.e. the starting row, from
4152 the specified string.
4154 Each non-whitespace character in the string is considered an alive
4155 cell, a newline will terminate the row, and further characters in the
4156 string will be ignored.
4159 Set the video rate, that is the number of frames generated per second.
4162 @item random_fill_ratio, ratio
4163 Set the random fill ratio for the initial cellular automaton row. It
4164 is a floating point number value ranging from 0 to 1, defaults to
4167 This option is ignored when a file or a pattern is specified.
4169 @item random_seed, seed
4170 Set the seed for filling randomly the initial row, must be an integer
4171 included between 0 and UINT32_MAX. If not specified, or if explicitly
4172 set to -1, the filter will try to use a good random seed on a best
4176 Set the cellular automaton rule, it is a number ranging from 0 to 255.
4177 Default value is 110.
4180 Set the size of the output video.
4182 If @option{filename} or @option{pattern} is specified, the size is set
4183 by default to the width of the specified initial state row, and the
4184 height is set to @var{width} * PHI.
4186 If @option{size} is set, it must contain the width of the specified
4187 pattern string, and the specified pattern will be centered in the
4190 If a filename or a pattern string is not specified, the size value
4191 defaults to "320x518" (used for a randomly generated initial state).
4194 If set to 1, scroll the output upward when all the rows in the output
4195 have been already filled. If set to 0, the new generated row will be
4196 written over the top row just after the bottom row is filled.
4199 @item start_full, full