/Zero-DCE-PyTorch

Pytorch implementation of the paper: Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement.

Primary LanguageJupyter Notebook

Zero-DCE: Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement

Pytorch implementation of Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement

Model Architecture

Result (Input--> Output)






References

[1] Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement - CVPR 2020 link

Citation

    @misc{guo2020zeroreference,
    title={Zero-Reference Deep Curve Estimation for Low-Light Image Enhancement},
    author={Chunle Guo and Chongyi Li and Jichang Guo and Chen Change Loy and Junhui Hou and Sam Kwong and Runmin Cong},
    year={2020},
    eprint={2001.06826},
    archivePrefix={arXiv},
    primaryClass={cs.CV}
}

Note

Some changes were made as compared to the original paper.

  • Learning rate was kept to 1e-3 instead of 1e-4.
  • Also, in the paper, equation for the curves is given as: LEn(x) = LEn−1(x) + αnLEn−1(x)(1 − LEn−1(x)) , whereas in the author's original implementation, the curve equation is taken as: LEn(x) = LEn−1(x) + αnLEn−1(x)(LEn−1(x) - 1)