sh run_train_generator.sh

Python val.py --pre_model './model/netG_model_epoch_50_iter_0.pth'

The performance on the 1-fold of CASIA NIR-VIS 2.0 dataset after running the above code:

@article{fu2021dvg,
  title={DVG-face: Dual variational generation for heterogeneous face recognition},
  author={Fu, Chaoyou and Wu, Xiang and Hu, Yibo and Huang, Huaibo and He, Ran},
  journal={IEEE TPAMI},
  year={2021}
}
@inproceedings{fu2019dual,
  title={Dual Variational Generation for Low-Shot Heterogeneous Face Recognition},
  author={Fu, Chaoyou and Wu, Xiang and Hu, Yibo and Huang, Huaibo and He, Ran},
  booktitle={NeurIPS},
  year={2019}
}
@article{fu2022towards,
  title={Towards Lightweight Pixel-Wise Hallucination for Heterogeneous Face Recognition},
  author={Fu, Chaoyou and Zhou, Xiaoqiang and He, Weizan and He, Ran},
  journal={IEEE TPAMI},
  year={2022}
}
@inproceedings{duan2020cross,
  title={Cross-spectral face hallucination via disentangling independent factors},
  author={Duan, Boyan and Fu, Chaoyou and Li, Yi and Song, Xingguang and He, Ran},
  booktitle={CVPR},
  year={2020}
}