[XviD-devel] 3 warp point GME

Christoph Lampert chl at math.uni-bonn.de
Sun Jun 29 14:17:47 CEST 2003


How to speed up 3 wp GME?

a) save first pass data for second. GME should not depend on 
   quantizer as block vectors do. 

b) during gradient decent, only regenerate blocks in which really
   something changes. For high res this might have a huge effect (or maybe 
   not, I'll have to check how many blocks really are affected.)

c) alternative: For every pel calculate difference to previous 
   step and update SAD only if diff != (0,0). 

d) ASM for innermost loop? 

e) algorithm with higher order of convergence? calculating the gradient is
   _very_ expensive at the moment, 6 full GMC-pictures are needed!

f) use fewer blocks, really identify "background".  But my first
   tests show, that it's difficult to find the right blocks.

g) just drop gradient decent, use something else?

gruel




On Sat, 28 Jun 2003, Christoph Lampert wrote:
> Hi,
> 
> I'll commit 3 warp point GME now. It tried to change as little as
> possible, so if there's a bug it's easy to go back. 
> 
> GME is functional (I hope) and I had good results for zooming scenes. 
> It's not optimized for speed, yet. Just a simple gradient decent, but in 
> 6 dimensional parameter space instead of 2 dimensioanl image space, so
> it's a little slow...
> 
> Please test with and without GME, since I had to change some other parts
> as well. I hope I didn't break anything...
> 
> gruel
> 
> P.S. Oh, there are two new files: motion/gmc.c and motion/gmc.h
> I guess someone has to add them to windows project files. 
> 
> -------------------------------------------------
> 
> Tempete CIF 512kbps twopass with different quality levels:
> 
> GMC with 3 point GME:
> fps = 38.92, length = 2497, psnr y = 28.99, u = 33.41, v = 35.63
> fps = 37.12, length = 2502, psnr y = 29.00, u = 33.53, v = 35.79
> fps = 33.12, length = 2480, psnr y = 29.22, u = 33.57, v = 35.85
> fps = 29.90, length = 2478, psnr y = 29.25, u = 33.60, v = 35.87
> 
> no GMC:
> fps = 99.16, length = 2487, psnr y = 28.83, u = 33.11, v = 35.31
> fps = 88.78, length = 2493, psnr y = 28.83, u = 33.22, v = 35.47
> fps = 71.03, length = 2478, psnr y = 29.04, u = 33.31, v = 35.60
> fps = 57.98, length = 2476, psnr y = 29.11, u = 33.35, v = 35.60
> 
> 
> _______________________________________________
> XviD-devel mailing list
> XviD-devel at xvid.org
> http://list.xvid.org/mailman/listinfo/xvid-devel
> 
> 



More information about the XviD-devel mailing list