Technically a remake of #12579

Experimenting with using Stable Diffusion to do a stylized comic format. I wanted to see what kind of quality was possible by pushing the tech, especially since it seems like other people haven't attempted it before (Not quite like this anyway). Backgrounds and the character were all generated separately from each other and then composited together via editing. Doing it that way gives me a lot more control over the scene composition than if I just generated each panel all at once. Mid-transformation segments were accomplished via inpainting and manual edits. Speech Bubbles, symbols, post-processing, and other effects were all manually edited in.

Might turn into a bigger thing, but this one page took long enough as is. The generation process eats up a lot of time, especially when it comes to maintaining consistency between shots.