clovaai / stargan-v2-tensorflow

StarGAN v2 - Official Tensorflow Implementation (CVPR 2020)

StarGAN v2 — Official TensorFlow Implementation [Paper] [Pytorch]

Implemented by Junho Kim

Requirements

• Tensorflow == 2.1.0
• Tensorflow-addons == 0.9.1
• opencv-python
• Pillow
• tqdm

Usage

├── dataset
   └── YOUR_DATASET_NAME
       ├── train
           ├── domain1 (domain folder)
               ├── xxx.jpg (domain1 image)
               ├── yyy.png
               ├── ...
           ├── domain2
               ├── aaa.jpg (domain2 image)
               ├── bbb.png
               ├── ...
           ├── ...
           
       ├── test
           ├── ref_imgs (domain folder)
               ├── domain1 (domain folder)
                   ├── ttt.jpg (domain1 image)
                   ├── aaa.png
                   ├── ...
               ├── domain2
                   ├── kkk.jpg (domain2 image)
                   ├── iii.png
                   ├── ...
               ├── ...
               
           ├── src_imgs
               ├── src1.jpg 
               ├── src2.png
               ├── ...

Train

python main.py --dataset celebA-HQ_gender --phase train

Test

python main.py --dataset celebA-HQ_gender --phase test

Tensorflow results (100K)

Latent-guided synthesis

CelebA-HQ

AFHQ

Reference-guided synthesis

CelebA-HQ

AFHQ

License

The source code, pre-trained models, and dataset are available under Creative Commons BY-NC 4.0 license by NAVER Corporation. You can use, copy, tranform and build upon the material for non-commercial purposes as long as you give appropriate credit by citing our paper, and indicate if changes were made.

For business inquiries, please contact [e-mail].
For technical and other inquires, please contact [e-mail].
For questions about the tensorflow implementation, please contact [e-mail].

Citation

If you find this work useful for your research, please cite our paper:

@inproceedings{choi2020starganv2,
  title={StarGAN v2: Diverse Image Synthesis for Multiple Domains},
  author={Yunjey Choi and Youngjung Uh and Jaejun Yoo and Jung-Woo Ha},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  year={2020}
}