1. [Publications](/publications)
2. TurboEdit: Text-Based Image Editing Using Few-Step Diffusion Models
 
 # TurboEdit: Text-Based Image Editing Using Few-Step Diffusion Models

  ![Publication image](/sites/default/files/styles/wide/public/default_images/default.jpeg?itok=qUFsuJCP "Publication image")

 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.



 ## Authors



Gilad Deutch (Tel-Aviv University)

Rinon Gal (NVIDIA)

Daniel Garibi (Tel-Aviv University)

Or Patashnik (Tel-Aviv University)

Daniel Cohen-Or (Tel-Aviv University)

 

 

 ## Publication Date



Thursday, August 1, 2024

 

 ## Published in



[SIGGRAPH Asia 2024](https://arxiv.org/abs/2408.00735)

 

 ## Research Area



[Computer Graphics](/research-area/computer-graphics)

[Computer Vision](/research-area/computer-vision)

[Generative AI](/research-area/generative-ai)