Official implementation of "PH-Net: Semi-Supervised Breast Lesion Segmentation via Patch-wise Hardness", CVPR 2024.
Due to the confidentiality agreement in commercial cooperation, we only provide codes of core modules and the whole trainable models for the convenience of comparisons.
- python >= 3.6
- pytorch >= 1.10.2
If you find our code useful, please cite our paper.
@inproceedings{jiang2024ph,
title={PH-Net: Semi-Supervised Breast Lesion Segmentation via Patch-wise Hardness},
author={Jiang, Siyao and Wu, Huisi and Chen, Junyang and Zhang, Qin and Qin, Jing},
booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},
pages={11418--11427},
year={2024}
}
We thank U²PL for part of their codes.