The official implementation of WiNet (MICCAI2024) 
Thanks @Junyu for the preprocessed IXI data. We followed the same training, validation, and testing protocol.
#pip < 24.1
conda create -n myenv python=3.9
conda activate myenv
pip install -r requirements.txt
Model weight can be found in
WiNet_code/ckpts
#Training
nohup python tr_IR_3D_WiNet.py >> "./WiNet_diff_IXI.out"
#Testing
python infer_bilinear_WiNet.py
The jupyter notebook file is used to get the evaluation results.
WiNet_code/Quantitative_Results/analysis.ipynb
@article{cheng2024winet,
title={WiNet: Wavelet-based Incremental Learning for Efficient Medical Image Registration},
author={Cheng, Xinxing and Jia, Xi and Lu, Wenqi and Li, Qiufu and Shen, Linlin and Krull, Alexander and Duan, Jinming},
journal={arXiv preprint arXiv:2407.13426},
year={2024}
}@inproceedings{cheng2024winet,
title={WiNet: Wavelet-based Incremental Learning for Efficient Medical Image Registration},
author={Cheng, Xinxing and Jia, Xi and Lu, Wenqi and Li, Qiufu and Shen, Linlin and Krull, Alexander and Duan, Jinming},
booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
pages={761--771},
year={2024},
organization={Springer}
}We would like to acknowledge the Fourier-Net and TransMorph projects, from which we have adopted some of the code used in our work.