Create HDR photos

Dumpr - Photo Fun Image evolution What is this? A simulated annealing like optimization algorithm, a reimplementation of Roger Alsing's excellent idea. The goal is to get an image represented as a collection of overlapping polygons of various colors and transparencies. We start from random 50 polygons that are invisible. Fitness is a sum of pixel-by-pixel differences from the original image. This implementation is based on Roger Alsing's description, though not on his code. How does it look after some time? 50 polygons (4-vertex) ~15 minutes 644 benefitial mutations 6,120 candidates 88.74% fitness 50 polygons (6-vertex) ~15 minutes 646 benefitial mutations 6,024 candidates 89.04% fitness 50 polygons (10-vertex) ~15 minutes 645 benefitial mutations 5,367 candidates 87.01% fitness 50 polygons (6-vertex) ~45 minutes 1,476 benefitial mutations 23,694 candidates 93.35% fitness 50 polygons (6-vertex) ~60 minutes 1,595 benefitial mutations 28,888 candidates 93.46% fitness Does it work on all images? It depends, success varies.