[x264-devel] Efficiently encoding synthetic data

[Chris Waters]

Hi,

I want to encode a synthetically sequence of images into an H.264 video with
as little CPU as possible. By synthetically generated I mean that my
program is generating the images ­ so I already know a number of
characteristics of each frame. For example I know which parts of each frame
differ from the previous frame ­ i.e. the rectangle containing the pixels
that have changed from the previous frame.

I am trying to work out if there is a way to tell x264 that only pixels
within a certain rectangle have changed when encoding each frame. It seems
to me that if I could do this then a lot of computation could be saved
because it wouldn't be necessary to do anything related to motion
estimation. Basically the first frame in my video would be an I-frame and
then every subsequent frame would be a P-frame containing only the pixels in
the changed rectangle.

I am more concerned about reducing the CPU overhead than the size of the
resulting file.

Can anyone give me any hints on where to look in the x264 source code to do
something like this?

Thanks,

Chris.


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.videolan.org/pipermail/x264-devel/attachments/20111106/56d8580e/attachment.html>