[FFmpeg-devel] [RFC] H.264: Supporting Recovery Point SEIs

Mans Rullgard mans
Fri Dec 25 21:51:24 CET 2009



"Jason Garrett-Glaser" <darkshikari at gmail.com> wrote:

>x264 will soon be committing Periodic Intra Refresh, an extremely
>powerful feature for low-latency, high-resilience streaming.  This
>will involve using SEI Recovery Points as seek points in H.264 video
>files.  This already works partially with container formats like MKV
>and MP4 that flag keyframes.  However, there is the following, massive
>gotcha:
>
>The standard specifies that, upon seeking to an SEI Recovery Point,
>that the decoder shall not display the video until recovery_frames_cnt
>frames have been decoded.
>
>This is relied upon heavily in Periodic Intra Refresh.
>
>How should this be implemented in ffmpeg?  Should the decoder return 0
>for every frame decoded after seeking until recovery_frames_cnt is
>reached?

Seems reasonable to me.  This is related in concept to the convergence duration found in some audio codecs.  I'm not sure how well we support that, but there was some discussion once.  I think it would make sense to use the same field(s) for both where applicable.
-- 
Mans Rullgard
mans at mansr.com



More information about the ffmpeg-devel mailing list