Yesterday I enjoyed sharing an unexpected art discovery from my recent wanders, so here’s another one, which was also something I quite accidentally wandered into on one of those perfect days which, for many reasons, I didn’t want to end.

If you are looking for a good use case for machine learning, this is it: a team from Google used ML, and some black-and-white photos from old art books, to not only replicate some priceless artworks which a bunch of Nazi arseholes destroyed, but to approximate the exact colours the artist would have used.

(There’s a lot of “if you know you know” here; only the best people do.)

Click on any image for full size.

You can read the full story of the restoration project (and see far better photos; I am definitely keeping my day job) here.

For a deep dive into the artist, his works, his times, and how all of that was turned into ashes literally and figuratively, read The Lady in Gold: The Extraordinary Tale of Gustav Klimt’s Masterpiece, Portrait of Adele Bloch-Bauer by Anne-Marie O’Connor.