Source: Dave ML
Summary:
Google reveals Self-guidance. It’s a zero-shot method that guides the internals of a diffusion model for greater control over image generation. With self-guidance, you can move or resize objects, or even replace them with items from real images, without changing the rest of the scene. You can also borrow the appearance of an entire other image or rearrange scenes to fit a desired composition.
Reading Time: 4 minutes
