[Sumover-dev] [svn commit] r4879 - vic/branches/mpeg4

[sumover-dev at cs.ucl.ac.uk]

Author: piers
Date: Wed Sep  1 11:38:16 2010
New Revision: 4879

Modified:
   vic/branches/mpeg4/CHANGES.html
   vic/branches/mpeg4/module.h

Log:
Updated to reflect info on latest changes
Module.h - corrected comment text on YUV format used by vic


Modified: vic/branches/mpeg4/CHANGES.html
==============================================================================
--- vic/branches/mpeg4/CHANGES.html	(original)
+++ vic/branches/mpeg4/CHANGES.html	Wed Sep  1 11:38:16 2010
@@ -32,8 +32,11 @@
       <li>Can also use the free GraphStudio from <a href="http://blog.monogram.sk/janos/tools/monogram-graphstudio/">http://blog.monogram.sk/janos/tools/monogram-graphstudio/</a>
     </ul>
   </ul>
-  <li>New grabber that uses the native API from the Blackmagic Design DeckLink SDK for Windows, MacOS and Linux
-  <li>libv4l support added for the V4L & V4L2 grabbers
+  <li>OSX now Leopard support (ie Activates 32bit build (-arch i386) as VIC uses deprecated Quicktime Capture interface which are only available in 32bit)
+  <li>New grabber-file - which supports raw YUV video files as video source
+  <li>New grabber that uses the native API from the Blackmagic Design DeckLink SDK for Windows, MacOS and Linux from HitLabNZ, integrated by Doug
+  <li>libv4l support added for the V4L and V4L2 grabbers
+  <li>Added iocom patches to repository against svn ffmpeg
 </ul>
 <H3>v2.8ucl-1.4.0beta</H3><UL>
   <li> H.264 decode functionality for IOCOM's IG2 default (non-standard) H.264 streams. Currently required patches ffmpeg h264 decoder.

Modified: vic/branches/mpeg4/module.h
==============================================================================
--- vic/branches/mpeg4/module.h	(original)
+++ vic/branches/mpeg4/module.h	Wed Sep  1 11:38:16 2010
@@ -54,8 +54,8 @@
 	int layer_;
 };
 
-/* YUV Frame uses I420(fourcc code) representation - Each frame is
- * stored as firstly the Y(Intensity) at full resolution - ie WxH
+/* YUV Frame uses Planar format I420(fourcc code) representation - Each
+ * frame is stored: Firstly the Y(Intensity) at full resolution - ie WxH
  * Followed by U(Chrominance Cb) subsampled 2x2 i.e. W/2*H/2
  * Followed by V(Chrominance Cr) also subsampled 2x2 
  * Hence 12 bits/pixel and frame size = WxH+2*(W/2*H/2)