Scalable Video Coding (SVC) smoothes MPEG4 h.264
YouTube, YouTube, YouTube. They’re always making announcements, some guff, some not. But their latest annoucement – that they’ll be re-encoding their videos using an enhancement to the MPEG4 h.264 codec, called scalable video coding (SVC) – is seriously big news.
In simple terms, the idea behind encoding videos is you divide up the picture into blocks, and estimate the motion between the blocks. This allows you to store the difference between frames, and not the whole frame, which reduces the file size. MPEG4 h.264 is particularly wonderful at doing this, but when packets of information get delayed or lost while traveling through the Internet, you get small glitches in the picture. SVC to the rescue! Those clever boffins at Fraunhofer have devised a way to enhance the existing codec to accommodate these packet drops. And the best news is that it seems we won’t need to change the players to make it work.
Hopefully it won’t be too long before we implement SVC in the Kensei encoder…
