[FFmpeg-devel] [PATCH] avfilter: add deband filter
Paul B Mahol
onemda at gmail.com
Thu Jul 16 21:49:20 CEST 2015
Dana 15. 7. 2015. 16:01 osoba "James Darnley" <james.darnley at gmail.com>
> On 2015-07-15 00:06, Paul B Mahol wrote:
> > Signed-off-by: Paul B Mahol <onemda at gmail.com>
> > ---
> > doc/filters.texi | 35 ++++++
> > libavfilter/Makefile | 1 +
> > libavfilter/allfilters.c | 1 +
> > libavfilter/vf_deband.c | 312
> > 4 files changed, 349 insertions(+)
> > create mode 100644 libavfilter/vf_deband.c
> > diff --git a/doc/filters.texi b/doc/filters.texi
> > index 9f7b976..0a07735 100644
> > --- a/doc/filters.texi
> > +++ b/doc/filters.texi
> > @@ -3678,6 +3678,41 @@ Violent denoise using a block size of
> > dctdnoiz=15:n=4
> > @end example
> > + at section deband
> > +
> > +Remove banding artifacts from input video.
> > +It works by replacing banded pixels with average value of referenced
> > +
> > +The filter accepts the following options:
> > +
> > + at table @option
> > + at item 1thr
> > + at item 2thr
> > + at item 3thr
> > + at item 4thr
> > +Set banding detection threshold for each plane. Default is 0.02. Range
> > +which value can be selected is from 0.008 to 0.5.
> I would suggest dropping the second "from" or rewording the third
> sentence as "Valid range is 0.008 to 0.5."
> > +If difference between current pixel and reference pixel is less than
> > +it will be considered as banded.
> > +
> > + at item range, r
> > +Banding detection range in pixels. Default is 16. If positive, random
> > +from range 0 to set value will be picked. If negative, exact of
> > +will be used.
> > +The range defines square of four pixels around current pixel.
> "If positive, random number in range 0 to r will be used."
> "If negative, exact absolute value will be used."
> > + at item direction, d
> > +Set direction from which four pixel will be compared. If positive,
> > +direction from 0 to set direction will be picked. If negative, exact of
> > +absolute value will be picked.
> Similar to above. Also, what does the direction mean? It's a float and
> looks like it represents radians (there is a sine/cosine pair below).
> The range is -2*PI..2*PI (should that also be mentioned here?) but what
> does the 0 represent? Looking at the code 0 is up then increasing moves
> in a clockwise direction.
> > + at item blur
> > +If enabled current pixel is compared with average value of all four
> > +surrounding pixels. By default is enabled. If disabled current pixel is
> "The default is enabled."
> > diff --git a/libavfilter/vf_deband.c b/libavfilter/vf_deband.c
> > new file mode 100644
> > index 0000000..1fd05d1
> > --- /dev/null
> > +++ b/libavfilter/vf_deband.c
> > @@ -0,0 +1,312 @@
> > +/*
> > + * Copyright (c) 2015 Niklas Haas
> > + * Copyright (c) 2015 Paul B Mahol
> > + *
> > + * Permission is hereby granted, free of charge, to any person
obtaining a copy
> > + * of this software and associated documentation files (the
"Software"), to deal
> > + * in the Software without restriction, including without limitation
> > + * to use, copy, modify, merge, publish, distribute, sublicense,
> > + * copies of the Software, and to permit persons to whom the Software
> > + * furnished to do so, subject to the following conditions:
> > + *
> > + * The above copyright notice and this permission notice shall be
> > + * all copies or substantial portions of the Software.
> > + *
> > + * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
> > + * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
> > + * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT
> > + * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR
> > + * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE,
> > + * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
DEALINGS IN THE
> > + * SOFTWARE.
> > + */
> This is the MIT license, right? Do we specifically note these anywhere?
Nope, imho not needed.
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
More information about the ffmpeg-devel