Home

Awesome

LoLi-IEA: low-light image enhancement algorithm

Overview

Official PyTorch implementation of LoLi-IEA: a low-light image enhancement algorithm presented at the SPIE Optical Engineering + Applications 2023 conference, San Diego, California, United States.

LoLi_Architecture

Requirements

  1. opencv-python == 4.9.0.80
  2. scikit-image == 0.22.0
  3. numpy == 1.24.3
  4. torch == 2.3.0+cu118
  5. Pillow == 10.2.0
  6. tqdm == 4.65.0
  7. natsort == 8.4.0
  8. torchvision == 0.18.0+cu118

Inference

To test the model, follow these steps:

  1. Download the weights for the Pretrained Model and place them in the ./Models directory.

  2. Place your images to be enhanced in the ./1_Input directory.

  3. Run the code with the following command:

    python main.py
    
    
  4. The enhanced images will be saved in the ./2_Output directory.

Citation

If this work contributes to your research, we would appreciate it if you could cite our paper:

@inproceedings{perez2023loli,
  title={LoLi-IEA: low-light image enhancement algorithm},
  author={Perez-Zarate, Ezequiel and Ramos-Soto, Oscar and Rodr{\'\i}guez-Esparza, Erick and Aguilar, German},
  booktitle={SPIE Optical Engineering + Applications},
  volume={12675},
  pages={230--245},
  year={2023},
  organization={SPIE}
}