[XviD-devel] Global Motion Compensation

skal xvid-devel@xvid.org
29 Oct 2002 11:42:40 +0100


	Hi all,

On Mon, 2002-10-28 at 22:12, Christoph Lampert wrote:
> On Mon, 28 Oct 2002, Michael Militzer wrote:
> > > @Isibaar: I'll have to change some of the code you added for quarterpel,
> > > because GMC needs VOL type 2, too, so only using m_quarterpel won't
> > > work anymore.
> > > Those was in BitstreamWriteVOLHeader (or what it's called...). Do you
> > > remember any other changes in headers due to quarterpel?
> > 
> > No, just VOLHeader (all places which start with an if(quarterpel)...)
> 
> Strange, because I keep getting color bleeds from QPEL+GMC support
> (without any S-VOPs, just sprites support in VOL header).
> 

	oh, talking about qpel and header, I just happen to notice,
	in Momusys, that qpel mode impacts the length of resync
	markers by requiring the presence of 1 additional zero.
	(see for instance: combined_decode.c, around line 3730).
	This is not mentioned in the norm, so far I've noticed...


	bye,

		Skal

(Btw, for something completely different: I've found in the news
this morning (!) the following reference:
http://mountains.ece.umn.edu/~guille/inpainting.htm
just wondering if this could be used for predictions (not in MPEG4, of
course)...)