Drag Your GAN: Interactive Point-based Manipulation on the Generative Image Manifold
Xingang Pan
·
Ayush Tewari
·
Thomas Leimkühler
·
Lingjie Liu
·
Abhimitra Meka
·
Christian Theobalt
SIGGRAPH 2023 Conference Proceedings
## Requirements
If you have CUDA graphic card, please follow the requirements of [https://github.com/NVlabs/stylegan3](https://github.com/NVlabs/stylegan3).
Otherwise (for GPU acceleration on MacOS with Silicon Mac M1/M2, or just CPU) try the following:
```sh
cat environment.yml | \
grep -v -E 'nvidia|cuda' > environment-no-nvidia.yml && \
conda env create -f environment-no-nvidia.yml
conda activate stylegan3
# On MacOS
export PYTORCH_ENABLE_MPS_FALLBACK=1
```
## Download pre-trained StyleGAN2 weights
To download pre-trained weights, simply run:
```sh
sh scripts/download_model.sh
```
If you want to try StyleGAN-Human and the Landscapes HQ (LHQ) dataset, please download weights from these links: [StyleGAN-Human](https://drive.google.com/file/d/1dlFEHbu-WzQWJl7nBBZYcTyo000H9hVm/view?usp=sharing), [LHQ](https://drive.google.com/file/d/16twEf0T9QINAEoMsWefoWiyhcTd-aiWc/view?usp=sharing), and put them under `./checkpoints`.
Feel free to try other pretrained StyleGAN.
## Run DragGAN GUI
To start the DragGAN GUI, simply run:
```sh
sh scripts/gui.sh
```
This GUI supports editing GAN-generated images. To edit a real image, you need to first perform GAN inversion using tools like [PTI](https://github.com/danielroich/PTI). Then load the new latent code and model weights to the GUI.
You can run DragGAN Gradio demo as well:
```sh
python visualizer_drag_gradio.py
```
## Acknowledgement
This code is developed based on [StyleGAN3](https://github.com/NVlabs/stylegan3). Part of the code is borrowed from [StyleGAN-Human](https://github.com/stylegan-human/StyleGAN-Human).
## License
The code related to the DragGAN algorithm is licensed under [CC-BY-NC](https://creativecommons.org/licenses/by-nc/4.0/).
However, most of this project are available under a separate license terms: all codes used or modified from [StyleGAN3](https://github.com/NVlabs/stylegan3) is under the [Nvidia Source Code License](https://github.com/NVlabs/stylegan3/blob/main/LICENSE.txt).
Any form of use and derivative of this code must preserve the watermarking functionality showing "AI Generated".
## BibTeX
```bibtex
@inproceedings{pan2023draggan,
title={Drag Your GAN: Interactive Point-based Manipulation on the Generative Image Manifold},
author={Pan, Xingang and Tewari, Ayush, and Leimk{\"u}hler, Thomas and Liu, Lingjie and Meka, Abhimitra and Theobalt, Christian},
booktitle = {ACM SIGGRAPH 2023 Conference Proceedings},
year={2023}
}
```
