Awesome
SCNet
Underwater Image Enhancement via Learning Water Type Desensitized Representations. Accepted by ICASSP 2022. [paper]
<div align=center><img src="img/1.png" height = "80%" width = "80%"/></div>You can run eval.py and to obtain the results using our pre-trained model [Baidu Drive(9xwn)] [Google Drive].
You can run get_performance.py to obtian the SSIM, PSNR and LPIPS scores.
To train the model, you need to prepare the dataset first [Baidu Drive(qwat)] [(Google Drive)]. Then, run main.py.
If you find SCNet is useful in your research, please consider citing our paper.
@INPROCEEDINGS{9747758,
author={Fu, Zhenqi and Lin, Xiaopeng and Wang, Wu and Huang, Yue and Ding, Xinghao},
booktitle={ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
title={Underwater Image Enhancement Via Learning Water Type Desensitized Representations},
year={2022},
pages={2764-2768},
doi={10.1109/ICASSP43922.2022.9747758}}