RUE-Net: Advancing Underwater Vision with Live Image Enhancement

Implementation of the paper RUE-Net: Advancing Underwater Vision with Live Image Enhancement
Here is the link to the PDF

Introduction

In this paper, we put forward a real-time underwater image enhancement network (RUE-Net). Compared to previous advanced models using single branch structure or traditional U-shaped networks, it consists of parallel Receptive Field Enhancement (RFE) Module and Fine Grain Detail (FGD) Module, which perform parallel modeling and fusion of image information at global and local scales.

Model structure

Recommended environment

Python 3.8
torch 1.8.0+cu111
pytorch-ssim

Train the Model

python training.py

Test the Model

python test.py

Citation

If you find our work useful, please consider citing the paper.

@article{wang2024rue,
  title={RUE-Net: Advancing Underwater Vision with Live Image Enhancement},
  author={Wang, Guocun and Chen, Chen and Xu, Hongli and Ru, Jingyu and Wang, Shuai and Wang, Zhenglong and Liu, Zhaofeng},
  journal={IEEE Transactions on Geoscience and Remote Sensing},
  year={2024},
  publisher={IEEE}
}

Acknowledgement

We are very grateful for the excellent work Shallow-UWnet, which has provided the basis for our framework.