4 ms·
> High quality results. When images are compressed or decompressed through Jpegli, more precise and psychovisually effective computations are performed and imag
by simonw 3y ago
> High quality results. When images are compressed or decompressed through Jpegli, more precise and psychovisually effective computations are performed and images will look clearer and have fewer observable artifacts.
Does anyone have a link to any example images that illustrate this improvement? I guess the examples would need to be encoded in some other lossless image format so I can reliably view them on my computer.
- n2d4 3y agoYou can find them in the mucped23.zip file linked here (encoded as PNG): https://github.com/google-research/google-research/tree/master/mucped23#results https://github.com/google-research/google-research/tree/mast...
- simonw 3y agoThanks - I downloaded that zip file (460MB!) and extracted one of the examples into a Gist: https://gist.github.com/simonw/5a8054f18f9ea3c560b628b16b00f32e https://gist.github.com/simonw/5a8054f18f9ea3c560b628b16b00f... Here's an original: https://gist.githubusercontent.com/simonw/5a8054f18f9ea3c560b628b16b00f32e/raw/4dba2c12cde6b9d1f5afccff36fccc2ad57fcbb7/originals-1624487.png https://gist.githubusercontent.com/simonw/5a8054f18f9ea3c560... And the jpegli-q95- version: https://gist.githubusercontent.com/simonw/5a8054f18f9ea3c560b628b16b00f32e/raw/4dba2c12cde6b9d1f5afccff36fccc2ad57fcbb7/jpegli-q95-yuv444-1624487.png https://gist.githubusercontent.com/simonw/5a8054f18f9ea3c560... And the same thing with mozjpeg-a95 https://gist.githubusercontent.com/simonw/5a8054f18f9ea3c560b628b16b00f32e/raw/4dba2c12cde6b9d1f5afccff36fccc2ad57fcbb7/mozjpeg-q95-yuv444-1624487.png https://gist.githubusercontent.com/simonw/5a8054f18f9ea3c560...
- tedunangst 3y agoWhat are the file sizes for those two?
- edflsafoiewq 3y agoEdit: I'm dumb.
- tedunangst 3y agoI would hope the jpegs compress better than png does.
- simonw 3y agoThe zip file doesn't have the originals, just the PNGs.
- IshKebab 3y agoThey're far too high quality to tell anything. There's no point comparing visually lossless images (inb4 "I am amazing and can easily tell...").
- deanresin 3y agoRight? I had all 3 open and quickly flipped over them saw no difference. Maybe I'm just uncultured.
- modeless 3y agoYou shouldn't compare the same quality setting across encoders as it's not standardized. You have to compare based on file size.
- pseudosavant 3y agoPerhaps try quality settings in the 70 range, and comparable output file sizes. 95 will be high-quality by definition.
- simonw 3y agoThere are some 70s and 65s in the gist: https://gist.github.com/simonw/5a8054f18f9ea3c560b628b16b00f32e https://gist.github.com/simonw/5a8054f18f9ea3c560b628b16b00f...
- masfuerte 3y agoYou said linked but like a fool I went looking for a zip in the repository. This is the link: https://cloudinary.com/labs/cid22/mucped23.zip https://cloudinary.com/labs/cid22/mucped23.zip (460MB)
- n2d4 3y agoI can't blame you, my comment originally didn't have the word "linked", I edited that in after I realized the potential misunderstanding. Maybe you saw it before the edit. My bad.
- masfuerte 3y agoHa ha! No worries. I thought it had changed but I frequently skim read and miss things so I wasn't sure.
- andrewla 3y agoAs an aside, jpeg is lossless on decode -- once encoded, all decoders will render the same pixels. Since this library produces a valid jpeg file, it should be possible to directly compare the two jpegs.
- JyrkiAlakuijala 3y agoIt is approximately correct. The rendering is standards compliant without pixel perfection and most decoders make different compromises and render slightly different pixels.
- nigeltao 3y ago> all decoders will render the same pixels Not true. Even just within libjpeg, there are three different IDCT implementations (jidctflt.c, jidctfst.c, jidctint.c) and they produce different pixels (it's a classic speed vs quality trade-off). It's spec-compliant to choose any of those. A few years ago, in libjpeg-turbo, they changed the smoothing kernel used for decoding (incomplete) progressive JPEGs, from a 3x3 window to 5x5. This meant the decoder produced different pixels, but again, that's still valid: https://github.com/libjpeg-turbo/libjpeg-turbo/commit/6d91e950c871103a11bac2f10c63bf998796c719 https://github.com/libjpeg-turbo/libjpeg-turbo/commit/6d91e9...
- iggldiggl 3y agoAnd nowadays, for subsampled images libjpeg post classic version 6 insists on doing the chroma upscaling using DCT where possible, so for classic 4:2:0 subsampled images (i.e. chroma resolution is half the luma resolution both horizontally and vertically) each subsampled 8x8 chroma block is now upscaled individually to 16x16 for the final image, which can and does introduce additional artefacts at the boundaries between each 16x16 px block in the final image. But the current libjpeg maintainer insists on that new algorithm because it is mathematically more beautiful… Granted, the introduced artefacts aren't massive, but under certain circumstances they are noticeable, which is how I stumbled across that topic in the first place. Thankfully, most software that isn't still stuck on libjpeg 6 has switched to libjpeg-turbo or some other library instead which continues using a more sensible algorithm for chroma upscaling.
- JyrkiAlakuijala 3y agohttps://twitter.com/jyzg/status/1622890389068718080 https://twitter.com/jyzg/status/1622890389068718080 Some earlier results. Perhaps these were with XYB color space, I don't remember ...