Spatial Decomposition Layer Segmentation network.

This repository contains a reference PyTorch implementation of SD-LayerNet, presented in SD-LayerNet: Semi-supervised retinal layer segmentation in OCT using disentangled representation with anatomical priors.

The source-code is currently provided as a reference, and cannot be trained out of the box, however, it contains the full implementation of SD-LayerNet (in the folder architecture), as well es the training code in model.py.

• PyTorch 1.9.1
• Cuda 11
• Python 3.9
• opencv-python

Soon we will add code for training on public datasets.