Diffusion Self-Guidance for Controllable Image Generation
This provides greater control over image generation for users of generative models, though it is incremental as it builds on existing guidance techniques.
The paper tackles the problem of controlling image generation in diffusion models beyond text descriptions by introducing self-guidance, a method that uses internal model representations to steer sampling for tasks like object manipulation and editing, achieving results such as modifying object positions and merging appearances without additional training.
Large-scale generative models are capable of producing high-quality images from detailed text descriptions. However, many aspects of an image are difficult or impossible to convey through text. We introduce self-guidance, a method that provides greater control over generated images by guiding the internal representations of diffusion models. We demonstrate that properties such as the shape, location, and appearance of objects can be extracted from these representations and used to steer sampling. Self-guidance works similarly to classifier guidance, but uses signals present in the pretrained model itself, requiring no additional models or training. We show how a simple set of properties can be composed to perform challenging image manipulations, such as modifying the position or size of objects, merging the appearance of objects in one image with the layout of another, composing objects from many images into one, and more. We also show that self-guidance can be used to edit real images. For results and an interactive demo, see our project page at https://dave.ml/selfguidance/