SDNet

Code of SDNet: A Versatile Squeeze-and-Decomposition Network for Real-Time Image Fusion.

@article{zhang2021sdnet,
  title={SDNet: A Versatile Squeeze-and-Decomposition Network for Real-Time Image Fusion},
  author={Zhang, Hao and Ma, Jiayi},
  journal={International Journal of Computer Vision},
  pages={1--25},
  year={2021},
  publisher={Springer}
}

Overview :

In this work, a squeeze-and-decomposition network (SDNet) is proposed to realize multi-modal and digital photography image fusion in real time. The proposed network architecture and loss functions are applicable to multiple fusion tasks.

running environment :

python=2.7, tensorflow-gpu=1.9.0.

To run :

Please follow the "readme.txt" in each image fusion task to prepare data, train and test.