📖 Towards Ultra-High-Definition Image Deraining: A Benchmark and An Efficient Method

Hongming Chen, Xiang Chen, Chen Wu, Zhuoran Zheng, Jinshan Pan, and Xianping Fu

Welcome to visit our website (专注底层视觉领域的信息服务平台) for low-level vision: https://lowlevelcv.com/

---

Setup

Type the command:

pip install -r requirements.txt

4K-Rain13k Dataset

(The datasets are hosted on both Google Drive and BaiduPan)

Download Link	Description
Google Drive / Baidu Netdisk	A total of 12,500 pairs for training and 500 pairs for testing.

Training and Testing

1. Please download the corresponding datasets and put them in the folder data/.
2. Follow the instructions below to begin training our model.

python train.py

3. Follow the instructions below to begin testing our model.

python test.py

Run the script then you can find the output visual results in the folder output/.

Evaluation

The PSNR, SSIM and MSE results are computed by using this Python Code.

Visual Results

Method	Download Link
LPNet	Google Drive / Baidu Netdisk
JORDER-E	Google Drive / Baidu Netdisk
RCDNet	Google Drive / Baidu Netdisk
SPDNet	Google Drive / Baidu Netdisk
IDT	Google Drive / Baidu Netdisk
Restormer	Google Drive / Baidu Netdisk
DRSformer	Google Drive / Baidu Netdisk
UDR-S2Former	Google Drive / Baidu Netdisk
UDR-Mixer	Google Drive / Baidu Netdisk

Citation

If you find this project useful in your research, please consider citing:

@article{chen2024towards,
  title={Towards Ultra-High-Definition Image Deraining: A Benchmark and An Efficient Method},
  author={Chen, Hongming and Chen, Xiang and Wu, Chen and Zheng, Zhuoran and Pan, Jinshan and Fu, Xianping},
  journal={arXiv preprint arXiv:2405.17074},
  year={2024}
}

Disclaimer

Please only use the dataset for research purposes.

Contact

If you have any questions, please feel free to reach me out at [email protected]