Abstract

We present a method for zero-shot, text-driven appearance manipulation in natural images and videos. Specifically, given an input image or video and a target text prompt, our goal is to edit the appearance of existing objects (e.g., object's texture) or augment the scene with new visual effects (e.g., smoke, fire) in a semantically meaningful manner. Our framework trains a generator using an internal dataset of training examples, extracted from a single input (image or video and target text prompt), while leveraging an external pre-trained CLIP model to establish our losses. Rather than directly generating the edited output, our key idea is to generate an edit layer (color+opacity) that is composited over the original input. This allows us to constrain the generation process and maintain high fidelity to the original input via novel text-driven losses that are applied directly to the edit layer. Our method neither relies on a pre-trained generator nor requires user-provided edit masks. Thus, it can perform localized, semantic edits on high-resolution natural images and videos across a variety of objects and scenes.  

Semi-Transparent Effects

Text2LIVE successfully augments the input scene with complex semi-transparent effects without changing irrelevant content in the image.

External Coverage

[Two Minute Papers]

Paper

Text2LIVE: Text-Driven Layered Image and Video Editing Omer Bar-Tal*, Dolev Ofri-Amar*, Rafail Fridman*, Yoni Kasten, Tali Dekel. (* indicates equal contribution) ECCV 2022 Oral. [paper]

 

Supplementary Material

[supplementary page]

 

Code

[code]

 

Bibtex

@inproceedings{bar2022text2live,
  title={Text2live: Text-driven layered image and video editing},
  author={Bar-Tal, Omer and Ofri-Amar, Dolev and Fridman, Rafail and Kasten, Yoni and Dekel, Tali},
  booktitle={European Conference on Computer Vision},
  pages={707--723},
  year={2022},
  organization={Springer}
}
			  

 

Acknowledgments

We thank Kfir Aberman, Lior Yariv, Shai Bagon, and Narek Tumanyan for their insightful comments. We thank Narek Tumanyan for his help with the baselines comparison.