First per-image PCA decomposition of Kodak suite reveals deliberate curation

github.com

9 points·PearsonZero··8 comments

These are the images: https://github.com/MohamedBakrAli/Kodak-Lossless-True-Color-...

Can someone explain what this is like I'm the idiot I am?

The Kodak Lossless True Color Image Suite is an ancient set of reference images from days of film used to make jpeg, and everything else, reproduce colors (and other thing) accurately. The linked GitHub repo is a PCA (Principal Component Analysis) decomposition of those images, which is a bunch of math (linear algebra) you'd use to make compressed representations of those pictures aka how you'd save them as a jpeg. (PCA does a lot of other stuff, but that's what it's doing here.)

Actual title -

Per-image principal component decomposition of the Kodak Lossless True Color Image Suite

From HN Guidelines [0]

Use the original title, unless it is misleading or linkbait; don't editorialize.

[0]: https://news.ycombinator.com/newsguidelines.html

I'd love some sort of "so what?" explainer in the README.

Or an abstract actually stating a conclusion versus the editorialized title in the HN post.

Definitely. This breaks HN guidelines for the title.

I don’t know where you read to find the conclusion that the editorialized title claims.

What is “curation” anyway? No kidding, Kodak curated the collection. The HN title is either unsubstantiated or useless.

I imagine that would be found in the associated paper, but I’m not sure if it’s been published yet. I’m having trouble finding it.

Parent Paper: Baetzel, J. (2026). Statistical Characterization of Inter-Channel Redundancy Structure in the Kodak Lossless True Color Image Suite. Per-Image Principal Component Decomposition of PCD0992.

Reply on news.ycombinator.com