DDBF CVPR2024

The code of Dispel Darkness for Better Fusion: A Controllable Visual Enhancer based on Cross-modal Conditional Adversarial Learning

@inproceedings{zhang2024dispel,
  title={Dispel Darkness for Better Fusion: A Controllable Visual Enhancer based on Cross-modal Conditional Adversarial Learning},
  author={Zhang, Hao and Tang, Linfeng and Xiang, Xinyu and Zuo, Xuhui and Ma, Jiayi},
  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  pages={26487--26496},
  year={2024}
}

Recommended Environment:

• python = 2.7
• tensorflow-gpu = 1.9.0
• numpy = 1.15.4
• scipy = 1.2.0
• pillow = 5.4.1
• scikit-image = 0.13.1

After testing, the following configuration is also OK:

• python = 3.6
• tensorflow-gpu = 1.9.0
• numpy = 1.19.2
• scipy = 1.5.4
• pillow = 8.4.0
• scikit-image = 0.17.2

Prepare data :

• Put low-light images in the "Dataset/Test/Low-light/..." for testing the function of low-light enhancement
• Put multi-modal images in the "Dataset/Test/Fusion/..." for testing the complete enhancement and fusion capabilities of DDBF.

Testing :

• Run "CUDA_VISIBLE_DEVICES=X python evaluate_Enhance.py" to enhance the provided low-light images.
• Run "CUDA_VISIBLE_DEVICES=X python evaluate_DDBF.py" to enhance and fuse the provided multi-modal images.