HN Explorer

Show HN: Pixel Art Generator Using Genetic Algorithm (github.com)

24 points by ponta17 ↗ HN
This project is a simple but fun demonstration of a genetic algorithm applied to image generation. It starts with a population of random images and evolves them over generations to resemble a target image. The output is an animated GIF that shows the entire evolution process.

7 comments

[ 3.1 ms ] story [ 31.8 ms ] thread
Evidlo
How do you think your example would compare against a simple lowpass + downsampling?

I think for most complicated pixel art, "reimagining" the image is necessary to keep it visually appealing and discernible at low-res, like in the examples here:

https://en.wikipedia.org/wiki/Pixel_art

__loam
Pixel art isn't just bucketing a higher resolution image into a lower resolution. If you're going to use that term, you should probably have at least a basic understanding of the medium.
u5wbxrc3
Interesting approach! The title is misleading though. I do not understand how this relates to pixel art. Pixel art is placing pixel on a constrained canvas and choosing colors on a limited palette which this does not seem to be. Maybe this could be considered blurring effect or image reconstruction/approximation?
Bimos
It looks like an unconstrained least square problem to me, which is convex and naive. Similar projects like reconstructing an image with transparent triangulars seems more interesting.