Our code is based on PyTorch-GAN, many thanks for this great work.
The model is trained with torch1.1 and cuda10.0. The all training data and reconstruction results can be found here (password:fzfl).
If you use these code or results in your researches or works, please cite our paper as follow:
@article{zhao2021hyperspectral,
title={Hyperspectral image shadow compensation via cycle-consistent adversarial networks},
author={Zhao, Min and Yan, Longbin and Chen, Jie},
journal={Neurocomputing},
volume={450},
pages={61--69},
year={2021},
publisher={Elsevier}
}