On Thu, 8 Oct 2015 12:20:25 +0200, Hendrik Leppkes wrote: > On Thu, Oct 8, 2015 at 11:30 AM, Stefano Sabatini <[email protected]> wrote: > > On date Wednesday 2015-10-07 15:03:32 +0200, Jean Delvare encoded: > >> The original interpolation algorithm behaved poorly on the borders and > >> did not even guarantee continuity at the borders. For this reason, a > >> second interpolation/blending pass was required on the borders to make > >> them seamless. > >> > >> However, since the interpolation algorithm was improved in June 2013, > >> the border issues no longer exist. The new algorithm does guarantee > >> continuity at the borders, making the second pass useless. A larger > >> band always increases the cumulated interpolation error. In most cases > >> it also increases the average interpolation error, even though the > >> samples in the band are only partially interpolated. > >> > >> For this reason I would like to get rid of the "band" parameter. As a > >> first step, let's change its default value from 4 to 1 and document it > >> as deprecated. > >> > >> I have benchmarked this change on a combination of input sources and > >> realistic logo areas. Lowering the band value from 4 to 1 resulted in > >> 8 to 39 % less interpolation error per frame (or 1 to 34 % less > >> interpolation error per luma sample.) > >> > >> Signed-off-by: Jean Delvare <[email protected]> > >> --- > >> Changes since v1: > >> * Added #ifs so that the deprecated options are dropped automatically > >> on next major version of libavfilter (suggested by Stefano Sabatini) > >> > >> doc/filters.texi | 4 +++- > >> libavfilter/vf_delogo.c | 17 +++++++++++++++-- > >> 2 files changed, 18 insertions(+), 3 deletions(-) > > > > Thanks, applied. > > This seems to have broken FATE, ie: > http://fate.ffmpeg.org/report.cgi?time=20151008101706&slot=x86_32-mingw-w64-dll-windows-native
Oops, yes, sorry. The different result is intended. I meant to update the test result but forgot to do so. Patch coming. -- Jean Delvare SUSE L3 Support _______________________________________________ ffmpeg-devel mailing list [email protected] http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
