Jiang-maomao/flare-removal

MFDNet: Multi-Frequency Deflare Network for Efficient Nighttime Flare Removal

Yiguo Jiang, Xuhang Chen , Chi-Man Pun📮 , Shuqiang Wang, Wei Feng (📮 Corresponding Author)

University of Macau, SIAT CAS, Huizhou University, Tianjin University

In The Visual Computer

⚙️ Usage

Installation

git clone https://github.com/Jiang-maomao/flare-removal.git
cd flare-removal

Training

To train a model with your own data/model, you can edit the config/config.py and run the following codes.

For single GPU training:

python train.py

For multiple GPUs training:

accelerate config
accelerate launch train.py

If you have difficulties on the usage of accelerate, please refer to Accelerate.

Inference

You can use the deflare.ipynb

Evaluation

You can run the evaluate.py

💗 Acknowledgements

This work was supported in part by the Science and Technology Development Fund, Macau SAR, under Grants 0141/2023/RIA2 and 0193/2023/RIA3.

🛎 Citation

If you find our work helpful for your research, please cite:

Jiang, Y., Chen, X., Pun, CM. et al. MFDNet: Multi-Frequency Deflare Network for efficient nighttime flare removal. Vis Comput (2024). https://doi.org/10.1007/s00371-024-03540-x