English

TurboEdit: Text-Based Image Editing Using Few-Step Diffusion Models

Computer Vision and Pattern Recognition 2024-08-02 v1 Graphics

Abstract

Diffusion models have opened the path to a wide range of text-based image editing frameworks. However, these typically build on the multi-step nature of the diffusion backwards process, and adapting them to distilled, fast-sampling methods has proven surprisingly challenging. Here, we focus on a popular line of text-based editing frameworks - the ``edit-friendly'' DDPM-noise inversion approach. We analyze its application to fast sampling methods and categorize its failures into two classes: the appearance of visual artifacts, and insufficient editing strength. We trace the artifacts to mismatched noise statistics between inverted noises and the expected noise schedule, and suggest a shifted noise schedule which corrects for this offset. To increase editing strength, we propose a pseudo-guidance approach that efficiently increases the magnitude of edits without introducing new artifacts. All in all, our method enables text-based image editing with as few as three diffusion steps, while providing novel insights into the mechanisms behind popular text-based editing approaches.

Keywords

Cite

@article{arxiv.2408.00735,
  title  = {TurboEdit: Text-Based Image Editing Using Few-Step Diffusion Models},
  author = {Gilad Deutch and Rinon Gal and Daniel Garibi and Or Patashnik and Daniel Cohen-Or},
  journal= {arXiv preprint arXiv:2408.00735},
  year   = {2024}
}

Comments

Project page: https://turboedit-paper.github.io/

R2 v1 2026-06-28T18:01:06.836Z