Hyper-Convolution Networks for Biomedical Image Segmentation

Code for our WACV 2022 paper "Hyper-Convolution Networks for Biomedical Image Segmentation"[https://arxiv.org/abs/2105.10559]

Convolutional Kernels are generated by a hyper-network instead of independtly learned

The input to the hyper-network are the spatial coordinates of the kernels

To initiate training, run: python main.py