I have some questions about lossless compression for webp.

[peiwei zhang]

Hi there，
Recently I have just started learning lossless compression for webp. I don't understand the process of transform and entropy coding. I wonder if webp, in lossless compression, does the whole image have the same transform and entropy encoding? Or is it to block different areas of the image and adopt different transform and entropy coding forms? I was reading the EncoderAnalyze() function and didn't see a specific chunk, so I was skeptical. If it has a different transform and entropy encoding using different blocks, please let me know where I can see it and what the block size is. Thank you for your help.
Thanks again

[peiwei zhang]

And If the webp exists block in EncoderAnalyze(), when the image is encoding, does it use blocks?  

[Vincent Rabaud]

The transforms, as defined in the specification at https://chromium.googlesource.com/webm/libwebp/+/refs/tags/v1.3.2/doc/webp-lossless-bitstream-spec.txt, are defined on the whole image. There is a notion of blocks though which applies to sub transforms. E.g., there can be a color transform, with parameters that are different depending on the block.

--
You received this message because you are subscribed to the Google Groups "WebP Discussion" group.
To unsubscribe from this group and stop receiving emails from it, send an email to webp-discuss...@webmproject.org.
To view this discussion on the web visit https://groups.google.com/a/webmproject.org/d/msgid/webp-discuss/4c8697ba-1eb7-4194-a8be-af74c3dcefecn%40webmproject.org.