https://github.com/fallenshock/FlowEdit

Official implementation of the paper: "FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models"
https://github.com/fallenshock/FlowEdit

diffusion-models editing editing-image generative-model rectified-flow text-guided-image-editing text-guided-image-manipulation

Last synced: 2 months ago
JSON representation

Official implementation of the paper: "FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models"

• Host: GitHub
• URL: https://github.com/fallenshock/FlowEdit
• Owner: fallenshock
• License: mit
• Created: 2024-11-22T17:01:07.000Z (3 months ago)
• Default Branch: main
• Last Pushed: 2024-12-12T07:43:10.000Z (2 months ago)
• Last Synced: 2024-12-12T08:29:42.306Z (2 months ago)
• Topics: diffusion-models, editing, editing-image, generative-model, rectified-flow, text-guided-image-editing, text-guided-image-manipulation
• Language: Python
• Homepage: https://matankleiner.github.io/flowedit/
• Size: 20.7 MB
• Stars: 2
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
[![Zero-Shot Image Editing](https://img.shields.io/badge/zero%20shot-image%20editing-Green)]([https://github.com/topics/video-editing](https://github.com/topics/text-guided-image-editing))

[![Python](https://img.shields.io/badge/python-3.8+-blue?python-3670A0?style=for-the-badge&logo=python&logoColor=ffdd54)](https://www.python.org/downloads/release/python-38/)

![PyTorch](https://img.shields.io/badge/torch-2.0.0-red?PyTorch-%23EE4C2C.svg?style=for-the-badge&logo=PyTorch&logoColor=white)

# FlowEdit

[Project](https://matankleiner.github.io/flowedit/) | [Arxiv](https://arxiv.org/abs/2412.08629) | [Demo](https://huggingface.co/spaces/fallenshock/FlowEdit) | [ComfyUI](#comfyui-implementation-for-different-models)

### Official Pytorch implementation of the paper: "FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models"

![](imgs/teaser.png)

## Installation

1. Clone the repository

2. Install the required dependencies using `pip install torch diffusers transformers accelerate sentencepiece protobuf` 


	* Tested with CUDA version 12.4 and diffusers 0.31.0

## Running examples

Run editing with Stable Diffusion 3: `python run_script.py --exp_yaml SD3_exp.yaml`

Run editing with Flux: `python run_script.py --exp_yaml FLUX_exp.yaml`

## Usage - your own examples

* Upload images to `example_images` folder. 

* Create an edits file that specifies: (a) a path to the input image, (b) a source prompt, (c) target prompts, and (d) target codes. The target codes summarize the changes between the source and target prompts and will appear in the output filename. 


See `edits.yaml` for example.

* Create an experiment file containing the hyperparamaters needed for running FlowEdit, such as `n_max`, `n_min`. This file also includes the path to the `edits.yaml` file


See `FLUX_exp.yaml` for FLUX usage example and `SD3_exp.yaml` for Stable Diffusion 3 usage example. 


For a detailed discussion on the impact of different hyperparameters and the values we used, please refer to our paper.

Run `python run_script.py --exp_yaml `

## ComfyUI implementation for different models 

* [FLUX](https://github.com/logtd/ComfyUI-Fluxtapoz)

* [LTX Video](https://github.com/logtd/ComfyUI-LTXTricks)

* [HunyuanLoom](https://github.com/logtd/ComfyUI-HunyuanLoom)

Implemented by [logtd](https://x.com/logtdx/status/1869095838016012462?s=48&t=6Yj6BZKooDOmH_JWRWjtHg)

## License

This project is licensed under the [MIT License](LICENSE).

### Citation

If you use this code for your research, please cite our paper:

```

@article{kulikov2024flowedit,

	title = {FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models},

	author = {Kulikov, Vladimir and Kleiner, Matan and Huberman-Spiegelglas, Inbar and Michaeli, Tomer},

	journal = {arXiv preprint arXiv:2412.08629},

	year = {2024}

	}

```