Awesome
FlowEdit
Project | Arxiv | Demo | ComfyUI
Official Pytorch implementation of the paper: "FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models"
Installation
-
Clone the repository
-
Install the required dependencies using
pip install torch diffusers transformers accelerate sentencepiece protobuf
<br>- 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. <br> 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 theedits.yaml
file<br> SeeFLUX_exp.yaml
for FLUX usage example andSD3_exp.yaml
for Stable Diffusion 3 usage example. <br> 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 <path to your experiment yaml>
ComfyUI implementation for different models
Implemented by logtd
License
This project is licensed under the MIT 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}
}