[XviD-devel] Mode decision

[skal]

On Wed, 2003-03-26 at 12:02, Christoph Lampert wrote:
> On 26 Mar 2003, skal wrote:
> > > After all, there is not reason why INTER4V
> > > should make things _worse_ at all in any case, except for a bad
> > > implementation.  
> > 	or bad criterion: in your orignal mail, the figure you showed
> > 	displayed a strong overlap of blocks wrongly qualified as 4MV.
> > 	So there's *also* room for improving the final choice
> > 	sad16<sum_of_sad8...
> > 
> > 	Did I overlook sthg?
> 
> I meant "bad implementation" as "bad full inter4v encoding process,
> including bad search and/or a bad mode decision criterion". You are of
> course right. Current SAD based INTER/INTER4V decision sucks!

	Now that I think of it, an complete opposite idea would
	be: always perform the 4mv search no matter what. And if
	the 4 motion vectors deduced are not that different, then
	qualify the block as INTER (and maybe launch a 16x16
	refinement, even). If I'm not mistaking, you may only pay
	the overhead of "4*SAD8x8 - SAD16x16", if the dive-toward-
	optimal is of sensibly equal length. Don't you think?

Skal