This repository is the official PyTorch implementation of Medical Imaging with Deep Learning (MIDL) paper "Rotation-Scale Equivariant Steerable Filters" by Yilong Yang, Srinandan Dasmahapatra and Sasan Mahmoodi. This is a Convolutional Neural Network (CNN) model that is robust to rotation, scaling, and translation transformation. The implementation is partially based on SESN, RotDCF, DCF-Net, RST-CNN, and N-Jet. Please refer to our paper for official references.
@article{yang2023rotation,
title={Rotation-Scale Equivariant Steerable Filters},
author={Yang, Yilong and Dasmahapatra, Srinandan and Mahmoodi, Sasan},
journal={Medical Imaging with Deep Learning},
year={2023}
}
- torch 1.7.1
- torchvision 0.8.2
- numpy 1.20.1
- scipy 1.5.2
- matplotlib 3.3.2