[theora] NHW Project - good quality improvement & new -l14 very high compression quality setting

Raphael Canut nhwcodec at gmail.com
Wed Dec 12 17:36:52 UTC 2018


 Hello,

I am still progressing with very high compression and I have corrected a
processing, so now very high compression (-l10 to -l13 quality settings)
has really better precision and so better quality.

I have also released a first draft of a new -l14 very high compression
quality setting.

More at: http://nhwcodec.blogspot.com/

Just a remark, the entropy coding schemes start to be less efficient at
-l14 very high compression, and we can really improve them and save
bits.For example the coding of the wavelet DC coefficients, wavelet DC
image is far from optimal.I have 4 compression modes for 4 different data
nature for the wavelet DC image, but I just choose the overall best coding
mode for the whole wavelet DC image... this is not optimal, I can really
save bits if for example I split the wavelet DC image in 16x16 blocks, and
give the one appropriate mode to each block, adapted to its nature.

Again using paq8px compressor, and simply compressing the .nhw compressed
files, we see that we can save 2.5KB per .nhw file...

Any feedback on the new -l14 quality setting and on the quality
improvements is, as usual, very welcome!

Many thanks again!
Cheers,
Raphael
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.xiph.org/pipermail/theora/attachments/20181212/9af2f1d2/attachment.html>


More information about the theora mailing list