[FFmpeg-devel] [PATCH] QCELP postfilter

[Ronald S. Bultje]

Hi,

On Mon, Mar 29, 2010 at 6:34 PM, Ronald S. Bultje <rsbultje at gmail.com> wrote:
> I guess my patch does something wrong. ;-). I'll see if I can get ref
> minus postfilter as a comparison.

Correction. The ref decoder always turned off the postfilter, because
the patch from that mailinglist thread you pointed out never turned it
on. ;-). No wonder it got worse.

New results: http://people.gnome.org/~rbultje/cmp2.png
Note how the last arrow ref decoder+postfilter is being clipped,
whereas our results (plus postfilter) are unclipped (and thus better).

bash-3.2$ tests/tiny_psnr ~/Desktop/out-ref-{,no}pf.wav 2 0 44
stddev:  300.86 PSNR: 46.76 bytes:  8820800/  8820800
bash-3.2$ tests/tiny_psnr ~/Desktop/out-{us,ref-nopf}.wav 2 0 44
stddev:  447.74 PSNR: 43.31 bytes:  8820800/  8820800
bash-3.2$ tests/tiny_psnr ~/Desktop/out-{us,ref-pf}.wav 2 0 44
stddev:  319.49 PSNR: 46.24 bytes:  8820800/  8820800

This shows PSNR decreases between us and ref minus vs. plus postfilter.

bash-3.2$ tests/tiny_psnr ~/Desktop/out-{nopf,ref-pf}.wav 2 0 44
stddev:  405.09 PSNR: 44.18 bytes:  8820800/  8820800
bash-3.2$ tests/tiny_psnr ~/Desktop/out-{nopf,ref-nopf}.wav 2 0 44
stddev:  281.97 PSNR: 47.33 bytes:  8820800/  8820800

This shows that PSNR increases between original (no postfilter code)
and ref minus/plus postfilter.

Conclusion: I think my patch does the same as the ref decoder and a
quick look at the code (postfilt() in code/postfilt.c) confirms that
we're doing mostly the same thing.

OK to apply?

Ronald