Concept
The idea is to train the neural network not to draw a specific object, but to work in a particular artistic style—to transform any subject into a stained-glass panel reminiscent of Tiffany lamps.
I generated a base Stable Diffusion style anchor, duplicated it through img2img across 24 new concepts, and compiled a dataset of 30 square images—all original generations, so licensing issues are non-existent.
1/9

































