On Wed, 3 Jan 2024 at 15:44, Stefano Sabatini <stefa...@gmail.com> wrote:
> On date Tuesday 2024-01-02 22:08:54 +0000, Kieran Kunhya wrote: > [...] > > The OP wants to test an implementation they have and they should do it > > using the API, not using ffprobe to test their third party implementation > > of something. > > > > > > "Parsed by FFmpeg" sounds like you are proposing to extend the FFmpeg > > > API to parse some more information (about the SEI subtype?) and make > > > it available through some more specific SEI side data? > > > > > > > As the OP says, this SEI data is from "MISB ST 0604" which is something > > that could be implemented in FFmpeg and parsed properly like all the > other > > side data instead of just writing random binary data to the terminal. > > Fine, so you are proposing to extend the parser to put this > information in the side data. This makes sense to me and sounds a > better approach than the original proposal. > SEIs are reordered so they need to go through the decoder to be frame accurate. I don't believe the parser is enough. Kieran _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
