Redefining Art in 2026: From Sketch-Based Models to Full Image Generation

I developed a custom image generation system based on a neural network architecture known as a UNET. In simple terms, this type of model learns how to gradually transform noise into meaningful images by recognizing patterns such as shapes, edges, and textures. What makes this work different is that the model was designed specifically to data consisted only of my own personal photographs and images that are in the public domain (meaning they are free to use and do not have