3: the entropy coding is just generic. The probabilities are affected
On Sat, Dec 9, 2023 at 9:34 AM peiwei zhang <
zhangp...@gmail.com> wrote:
>
> Hi, I have another question. Because my current knowledge level is not enough, I would like to ask some conclusions about webp lossless compression. First, the lossless compression of webp is to transform before entropy encoding, right? If so, on what basis does it choose the mode of transform? Second, is the lossless compression selection entropy encoding for webp selected according to the crunch_configs_size obtained by EncoderAnalysz? The last question is, for lossless compression of webp, is there a correlation between choosing transform and choosing entropy encoding? Or should we first determine the transform mode and then choose the entropy encoding method according to the converted data? My code uses method = 6, quality = 100. Thank you so much. Thank you for your help!
>
> 在2023年12月6日星期三 UTC+8 00:18:06<Vincent Rabaud> 写道:
>>