Pretty clever. Somewhat related, if you want to match the colours of an entire image to an other, you can use optimal transport [1]<p>Unrelated, the author has a very nice take on generating the organic grid from Oskar Stålberg’s Townscaper game [2].<p>[1]: <a href="https://codimd.math.cnrs.fr/s/2eRBqV9zl" rel="nofollow">https://codimd.math.cnrs.fr/s/2eRBqV9zl</a><p>[2]: <a href="https://andersource.dev/2020/11/06/organic-grid.html" rel="nofollow">https://andersource.dev/2020/11/06/organic-grid.html</a>
A more pedestrian imagemagick script that uses a "fuzz value" that's a number of degrees on a hue wheel instead:<p><a href="http://www.fmwconcepts.com/imagemagick/replacecolor/index.php" rel="nofollow">http://www.fmwconcepts.com/imagemagick/replacecolor/index.ph...</a>
"Flowers replaced to red" looks like it also tinted the whole sky red. In general, all the image captions imply that the algorithm is able to identify objects (flowers, ice cream scoop, shirt) and only change that object's color.
Sorry for the noob question but reading through this I can't make sense of why this is the desired transformation:<p>"Desired transformation: σ(σ(IT)N)"<p>What is I, T, and N?
Also there is <a href="https://www.cs.huji.ac.il/~yweiss/Colorization/" rel="nofollow">https://www.cs.huji.ac.il/~yweiss/Colorization/</a><p>Had implemented a Khoros filter box based on this for my Computer Graphics class back then.