This repo contains the pytorch implementation of adversarial data augmentation, which supports to perform adversarial training on a chain of image photometric transformations and geometric transformations for improved consistency regularization. Please cite our work if you find it useful in your work
Under construction.
- matplotlib>=2.0
- seaborn>=0.10.0
- numpy>=1.13.3
- SimpleITK>=2.1.0
- skimage>=0.0
- torch>=1.9.0
- Install PyTorch and other required python libraries with:
pip install -r requirements.txt
- Play with the provided jupyter notebook to check the enviroments
Under construction
If you find this useful for your work, please consider citing
@INPROCEEDINGS{Chen_MICCAI_2020_Realistic,
title = "Realistic Adversarial Data Augmentation for {MR} Image
Segmentation",
booktitle = "Medical Image Computing and Computer Assisted Intervention --
{MICCAI} 2020",
author = "Chen, Chen and Qin, Chen and Qiu, Huaqi and Ouyang, Cheng and
Wang, Shuo and Chen, Liang and Tarroni, Giacomo and Bai, Wenjia
and Rueckert, Daniel",
publisher = "Springer International Publishing",
pages = "667--677",
year = 2020
}