FMR-Net: a fast multi-scale residual network for low-light image enhancement

[Paper]

Representitive Results

Training process

run 'train.py'

Testing process

We provide three pre-trained models for the LOL dataset, the FiveK dataset and the results of training them together. You can also train with your own dataset. Modify your pre-trained model path in 'data_test.py

If you find this work useful for you, please cite

@article{chen2024fmr,
  title={FMR-Net: a fast multi-scale residual network for low-light image enhancement},
  author={Chen, Yuhan and Zhu, Ge and Wang, Xianquan and Shen, Yuhuai},
  journal={Multimedia Systems},
  volume={30},
  number={2},
  pages={73},
  year={2024},
  publisher={Springer}
}

Contact

If you have any questions, please contact Yuhan Chen at cyh1217552389@gmail.com