Yannis Guyon

I confirm that `convert image1.png -alpha off image1.png` solves the issue for me, thanks Luca.

> ... so we need "-layers coalesce", or just "-coalesce". > > The input is 80x80 pixels, so resizing to 80x80 doesn't do much, but the output seems fine. As...

Here is a related discussion about Exif orientation import in libavif from Jpeg/PNG to irot/imir: https://github.com/AOMediaCodec/libavif/pull/1099#discussion_r970928340 Even though MIAF says `There should be no image transformations expressed by Exif`, the...

@dlemstra Hi, the problem seems to still exist: https://bugs.chromium.org/p/webp/issues/detail?id=628

I am not so sure I follow. You said: > We have removed the call to `-coalesce` and this issue should be resolved in the next release. But the original...

> I haven't had/made time to investigate this. That is fine. Could we reopen this issue in the meantime, to signal there is some unexpected behavior? Or is there another...

> This is now the expected behavior so I don't see a reason to reopen this issue. Understood. I meant "unexpected" as in "users expect a similar visual representation when...

> cc: @y-guyon for any feedback on the processing time, maybe there is something about the config that could be optimized I think so. This [comparison](https://storage.googleapis.com/demos.webmproject.org/webp/cmp/2024_03_25/index.html?batch=batches/avif_6.json&batch=batches/webp_4.json) uses libavif+aom instead of...

Hi all, I would like to suggest a precision to the next ["Media - Bits per pixel by format"](https://almanac.httparchive.org/en/2022/media#bits-per-pixel-by-format) section of the Web Almanac. I referred a lot to that...