On 3/4/22, Sebastian Wallkötter <sebast...@wallkoetter.net> wrote: > Hi! > > I'm trying to understand how ffmpeg goes about demosaicing and which Bayer > formats are available. Currently, I'm looking at the bayer_rggb8 format, and > what is confusing me is that the format info reports that the format has > less than 8 bits per component: > > https://github.com/FFmpeg/FFmpeg/blob/25c8507818d8559a6654a5b30a0f8aae11a48181/libavutil/pixdesc.c#L1942-L1966 > > However, when I look up how libswscale does the conversion, it treats the > format as having 8bit in each element of the Bayer mosaic: > > https://github.com/FFmpeg/FFmpeg/blob/1a502b99e818ee7b8b2b56c4f5c27e31f674c555/libswscale/bayer_template.c#L56-L70 > > How do these two things reconcile? I am quite certain that this is me > misunderstanding something, so a pointer on what I am mixing up would be > much appreciated. > > Many thanks in advance!
It is there as stub and meaningless. When debayering ffmpeg uses all bits in pattern. There are bunch of ways how to do debayering. > > Best, > Sebastian > _______________________________________________ > ffmpeg-user mailing list > ffmpeg-user@ffmpeg.org > https://ffmpeg.org/mailman/listinfo/ffmpeg-user > > To unsubscribe, visit link above, or email > ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe". > _______________________________________________ ffmpeg-user mailing list ffmpeg-user@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-user To unsubscribe, visit link above, or email ffmpeg-user-requ...@ffmpeg.org with subject "unsubscribe".