This is an official release of the paper XBound-Former: Toward Cross-scale Boundary Modeling in Transformers, including the network implementation and the training scripts.
[XBound-Former: Toward Cross-scale Boundary Modeling in Transformers]
Jiacheng Wang, Fei Chen, Yuxi Ma, Liansheng Wang, Zhaodong Fei, Jianwei Shuai, Xiangdong Tang, Qichao Zhou, Jing Qin
In: Transactions on Medical Imaging (TMI), 2023
[arXiv][Bibetex]
- [1/11 2022] This paper has been accepted to TMI.
- [5/27 2022] We have released the training scripts.
- [5/19 2022] We have created this repo.
- Network
- Pre-processing
- Training Codes
- Pretrained Weights
For more details or any questions, please feel easy to contact us by email ([email protected]).
Please download the dataset from ISIC challenge and PH2 website.
Please run:
$ python utils/resize.py
You need to change the File Path to your own and select the correct function.
Please run:
$ python src/train.py
You need to change the File Path to your own and select the correct function.
Download the pretrained weight for ISCI-2016&$ph^2$ dataset from Google Drive and move to the logger dir.
Then, please run:
$ python src/test.py
The ISIC-2016&$ph^2$ dataset:
If you find XBound-Former useful in your research, please consider citing:
@article{wang2023xbound,
title={XBound-Former: Toward Cross-scale Boundary Modeling in Transformers},
author={Wang, Jiacheng and Chen, Fei and Ma, Yuxi and Wang, Liansheng and Fei, Zhaodong and Shuai, Jianwei and Tang, Xiangdong and Zhou, Qichao and Qin, Jing},
journal={IEEE Transactions on Medical Imaging},
year={2023},
publisher={IEEE}
}
and the prior work, BAT, as:
@inproceedings{wang2021boundary,
title={Boundary-Aware Transformers for Skin Lesion Segmentation},
author={Wang, Jiacheng and Wei, Lan and Wang, Liansheng and Zhou, Qichao and Zhu, Lei and Qin, Jing},
booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
pages={206--216},
year={2021},
organization={Springer}
}