R2RNet
Official code of "R2RNet: Low-light Image Enhancement via Real-low to Real-normal Network." Jiang Hai, Zhu Xuan, Ren Yang, Yutong Hao, Fengzhu Zou, Fang Lin, and Songchen Han(Submitted to IEEE transaction on Image Processing)
Paper link: https://arxiv.org/abs/2106.14501
Network Architecture
Pytorch
This is a Pytorch implementation of R2RNet.
Requirements
- Python 3.x
- Pytorch 1.x.0
Dataset
You can download the LSRW dataset from: https://pan.baidu.com/s/1UxFllrtRSh4E8ir8LdTb9w (code: wmr1)
If you use our code and dataset, please cite our paper.
Pre-trained model
The pre-trained models can be download from: https://pan.baidu.com/s/1emGK8_JHNktoEn0dj5OnhQ (code: wmrr)
You shold download the VGG model (https://pan.baidu.com/s/1Rn2NwHt9eZgfg6hQP-DrlQ code:wmr1)and put it into ./model.
Testing Usage
python predict.py
Training Usage
python trian.py
Reference
Code borrows heavily from https://github.com/aasharma90/RetinexNet_PyTorch.
