AttentionGAN-v1-multi

This repository provides a PyTorch implementation of StarGAN + AttentionGAN-v1 for the Multi-Domain Image-to-Image Translation. StarGAN can flexibly translate an input image to any desired target domain using only a single generator and a discriminator.

Order: The Learned Attention Mask, The Learned Content Mask, Final Results.

Order: The Learned Attention Mask, The Learned Content Mask, Final Results.

Dependencies

• Python 3.5+
• PyTorch 0.4.0+
• TensorFlow 1.3+ (optional for tensorboard)

Usage

1. Cloning the repository

$ git clone https://github.com/Ha0Tang/AttentionGAN
$ cd AttentionGAN/
$ cd AttentionGAN-v1-multi/

2. Downloading the dataset

To download the CelebA dataset:

$ bash download_dataset.sh celeba

To download the RaFD dataset, you must request access to the dataset from the Radboud Faces Database website. Then, you need to create a folder structure as described here.

3. Training

To train StarGAN on CelebA, run the training script below. See here for a list of selectable attributes in the CelebA dataset. If you change the selected_attrs argument, you should also change the c_dim argument accordingly.

To train StarGAN on RaFD:

$ python main.py --mode train --dataset RaFD --image_size 128 --c_dim 8 \
                 --sample_dir stargan_rafd/samples --log_dir stargan_rafd/logs \
                 --model_save_dir stargan_rafd/models --result_dir stargan_rafd/results

To train StarGAN on CelebA:

$ python main.py --mode train --dataset CelebA --image_size 128 --c_dim 5 \
                 --sample_dir stargan_celeba/samples --log_dir stargan_celeba/logs \
                 --model_save_dir stargan_celeba/models --result_dir stargan_celeba/results \
                 --selected_attrs Black_Hair Blond_Hair Brown_Hair Male Young

For more training options, please refer to StarGAN.

4. Testing

To test StarGAN on RaFD:

$ python main.py --mode test --dataset RaFD --image_size 128 \
                 --c_dim 8 --rafd_image_dir data/RaFD/test \
                 --sample_dir stargan_rafd/samples --log_dir stargan_rafd/logs \
                 --model_save_dir stargan_rafd/models --result_dir stargan_rafd/results

To test StarGAN on CelebA:

$ python main.py --mode test --dataset CelebA --image_size 128 --c_dim 5 \
                 --sample_dir stargan_celeba/samples --log_dir stargan_celeba/logs \
                 --model_save_dir stargan_celeba/models --result_dir stargan_celeba/results \
                 --selected_attrs Black_Hair Blond_Hair Brown_Hair Male Young

5. Pretrained model

To download a pretrained model checkpoint, run the script below. The pretrained model checkpoint will be downloaded and saved into ./rafd_128_pretrained/models directory.

$ bash download_pretrained.sh rafd_128
$ bash download_pretrained.sh rafd_256
$ bash download_pretrained.sh celeba_128
$ bash download_pretrained.sh celeba_256

To translate images using the pretrained model, run the evaluation script below. The translated images will be saved into ./rafd_128_pretrained/results directory.

$ python main.py --mode test --dataset RaFD --image_size 128 --c_dim 8 \
                 --model_save_dir='rafd_128_pretrained/models' \
                 --result_dir='rafd_128_pretrained/results'

The model pretrained by the original StarGAN on RaFD dataset with 256 resolution can by downloaded by using,

$ bash download_pretrained.sh stargan_rafd_256 

Then use the original StarGAN code to generated results.

Citation

If you use this code for your research, please cite our papers.

@article{tang2019attentiongan,
  title={AttentionGAN: Unpaired Image-to-Image Translation using Attention-Guided Generative Adversarial Networks},
  author={Tang, Hao and Liu, Hong and Xu, Dan and Torr, Philip HS and Sebe, Nicu},
  journal={arXiv preprint arXiv:1911.11897},
  year={2019}
}

@inproceedings{tang2019attention,
  title={Attention-Guided Generative Adversarial Networks for Unsupervised Image-to-Image Translation},
  author={Tang, Hao and Xu, Dan and Sebe, Nicu and Yan, Yan},
  booktitle={International Joint Conference on Neural Networks (IJCNN)},
  year={2019}
}

Acknowledgments

This source code is inspired by StarGAN.

Contributions

If you have any questions/comments/bug reports, feel free to open a github issue or pull a request or e-mail to the author Hao Tang (hao.tang@unitn.it).