MIFFuse

This code is for "MIFFuse: A Multi-level Feature Fusion Network for Infrared and visible Images"

Requirements

1. Python 3.6

2. Pytorch 1.7.0

3. CUDA 11.0

4. The images needed for the training process can be downloaded here (only 10 sample images are included in the project)

Code Structure

• train: Training code on the dataset.
• test: Test the model on TNO/CVC_14 dataset
• test_RGB: Test the model on Flir dataset
• MIF_net: MIFFufuse architecture

Citation

@ARTICLE{MIFFuse,
  author={Zhu, Depeng and Zhan, Weida and Jiang, Yichun and Xu, Xiaoyu and Guo, Renzhong},
  journal={IEEE Access}, 
  title={MIFFuse: A Multi-Level Feature Fusion Network for Infrared and Visible Images}, 
  year={2021},
  volume={9},
  number={},
  pages={130778-130792},
  doi={10.1109/ACCESS.2021.3111905}}

If you have any question, please email to me zhudepeng@mails.cust.edu.cn