LiuZhu-CV / BDLFusion

Bi-level Dynamic Learning for Jointly Multi-modality Image Fusion and Beyond (IJCAI 23)

BDLFusion

Codes of Bi-level Dynamic Learning for Jointly Multi-modality Image Fusion and Beyond

Zhu Liu, Jinyuan Liu, Guanyao Wu, Long Ma, Xin Fan, Risheng Liu. In IJCAI 2023. Paper

Fusion Reuslts and Chinese Version

The source images and fused results on TNO, RoadScene, MFNet and M3FD (600 images) are provided in link

中文版介绍提供在此链接 link

Welcome all comparision and disscussion! If you have any questions, please sending an email to "[e-mail]"

Requirements

• Python 3.7
• PyTorch 1.10.1
• Checkpoint of detection Checkpoint

Usage

Testing

Run "python test.py" to test the model.

Training

Run "python train.py" to train the model.

The warmstart checkpoint and saliency weight maps are provided at url (https://drive.google.com/drive/folders/1WxxuXFDX4-18DfAJjDWoemREBDRcEaDH?usp=drive_link)

Workflow

Results of detection

Results of segmentation

Citation

If you use this code for your research, please cite our paper.

@article{liu2023bilevel,
  title={Bi-level Dynamic Learning  for Jointly Multi-modality Image Fusion and Beyond},
  author={Zhu Liu and Jinyuan Liu and Guanyao Wu and Long Ma and Xin Fan and Risheng Liu},
  journal={IJCAI},
  year={2023},
}