“Zero-shot Image-to-image Translation” by Parmar, Singh, Zhang, Li, Lu, et al. …
Conference:
Type(s):
Title:
- Zero-shot Image-to-image Translation
Session/Category Title: Image and Video Editing
Presenter(s)/Author(s):
Moderator(s):
Abstract:
We propose pix2pix-zero, a diffusion-based, image-to-image approach that allows for specifying the edit direction on the fly (e.g., cat to dog). Our method can directly use pre-trained text-to-image diffusion models, such as Stable Diffusion, for editing real and synthetic images while preserving the input image’s structure. Our method is training free and prompt free.