Diffusion-Based Conditional Image Editing through Optimized Inference with Guidance
This work addresses image editing for users needing precise control over generated content, but it is incremental as it builds on existing diffusion models with a novel guidance technique.
The paper tackles the problem of text-driven image-to-image translation by proposing a training-free method that uses a pretrained diffusion model to generate images aligned with a target prompt while preserving the source image's structure and background, achieving outstanding performance on various tasks.
We present a simple but effective training-free approach for text-driven image-to-image translation based on a pretrained text-to-image diffusion model. Our goal is to generate an image that aligns with the target task while preserving the structure and background of a source image. To this end, we derive the representation guidance with a combination of two objectives: maximizing the similarity to the target prompt based on the CLIP score and minimizing the structural distance to the source latent variable. This guidance improves the fidelity of the generated target image to the given target prompt while maintaining the structure integrity of the source image. To incorporate the representation guidance component, we optimize the target latent variable of diffusion model's reverse process with the guidance. Experimental results demonstrate that our method achieves outstanding image-to-image translation performance on various tasks when combined with the pretrained Stable Diffusion model.