/ACL-GAN

Unpaired Image-to-Image Translation using Adversarial Consistency Loss, ECCV 2020

Primary LanguagePythonOtherNOASSERTION

Paper

Yihao Zhao, Ruihai Wu, Hao Dong, "Unpaired Image-to-Image Translation using Adversarial Consistency Loss", ECCV 2020

arXiv: https://arxiv.org/abs/2003.04858

Code usage

For environment:

conda env create -f acl-gan.yaml

For training:

python train.py --config configs/male2female.yaml

For test:

python test.py --config configs/male2female.yaml --input inputs/test_male.jpg --checkpoint ./models/test.pth

Experimental Results

Ablation study

ablation_study

Male-to-female

male2female

Glasses Removal

glasses_removal

Selfie-to-anime

selfie2anime

For more results, please refer to our paper.

Code Base

Xun Huang, Ming-Yu Liu, Serge Belongie, Jan Kautz, "Multimodal Unsupervised Image-to-Image Translation", ECCV 2018

Citation

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

@inproceedings{zhao2020aclgan,
  title={Unpaired Image-to-Image Translation using Adversarial Consistency Loss},
  author={Zhao, Yihao and Wu, Ruihai and Dong, Hao},
  booktitle={ECCV},
  year={2020}
}