[FFmpeg-devel] Scalable Video Encoding and FGS

[Loren Merritt]

On Fri, 19 Oct 2007, Erik van Zijst wrote:
>
> I've considered layering on frame-type again (also comes up a lot in the
> research community), but I don't think it will be able to lower the
> bitrate significantly due to their relative sizes.
>
> In MPEG1 b-frames are roughly 70 times smaller than i-frames [1], so
> cutting the bitrate in half, requires removal of 70 b-frames per i-frame
> (sure, these figures old and mpeg1-specific, but what matters is the
> concept).
>
> So frame removal could be used to gain some scalability, but IMHO it
> cannot match the granularity of FGS and SVC.
>
> [1]
> "The typical data rate of an I-frame is 1 bit per pixel while that of a
> P-frame is 0.1 bit per pixel and for a B-frame, 0.015 bit per pixel." --
> http://bmrc.berkeley.edu/frame/research/mpeg/mpeg_overview.html

That't a factor of 7, not 70. And What matters is the ration between I+P 
and B framse, not just I and B.

--Loren Merritt