[XviD-devel] adaptive quant improvement

Marco "elcabesa" Belli elcabesa at inwind.it
Tue Feb 4 19:08:12 CET 2003


> The more common term is texture masking, texture masking is really just the
> result of local/spatial contrast masking over a larger area. Heavily
> textured/highly active parts of the image will have high contrasts between
> pixels, and as such will provide lots of contrast masking.
>
> A model incorporating both texture and luminance masking developed for JPEG
> which, with some adaption of the parameters, might work decently in the
> present (pre motion-compensation) adaptive quantization code is described
> in http://home.student.utwente.nl/m.f.al/00999032.pdf

thank you=)

to ed gomez
have you  tested my code? do you like it? does it work ? any idea about how to 
make it better?


More information about the XviD-devel mailing list