self-supervised-landmarks

Repository for self-supervised landmark discovery

Requirements

pytorch
pynrrd (for 3d images)

Usage

The use of this models is via config files, an example config file for Shepp-logan phantom dataset is given in ./configs/phantom_data.json

To train the model

python train.py --model="2d or 3d" --config_file="path to config file"

The network could be 2d or 3d and the second argument is the config file path. all the other parameters including the save and data director is inside teh config file

For inference

python test.py --model="2d or 3d" --config_file="path to config file" --redu_remove --use_best --num_out=num ts to be retained

redu_remove is a boolean argument that determines if redundant points are removed or not use best is also a boolean argument that determines if the best checkpoint is used or the final checkpoint. num_out is an integer that determines the number of particles to be retained after redundancy removal

Reference

If you are utilizing this code please cite one of the following

Leveraging unsupervised image registration for discovery of landmark shape descriptor

@article{bhalodia2021leveraging,
title = {Leveraging unsupervised image registration for discovery of landmark shape descriptor},
journal = {Medical Image Analysis},
volume = {73},
pages = {102157},
year = {2021},
issn = {1361-8415},
doi = {https://doi.org/10.1016/j.media.2021.102157},
url = {https://www.sciencedirect.com/science/article/pii/S1361841521002036},
author = {Riddhish Bhalodia and Shireen Elhabian and Ladislav Kavan and Ross Whitaker}
}

Self-supervised discovery of anatomical shape landmarks

@inproceedings{bhalodia2020self,
  title={Self-supervised discovery of anatomical shape landmarks},
  author={Bhalodia, Riddhish and Kavan, Ladislav and Whitaker, Ross T},
  booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
  pages={627--638},
  year={2020},
  organization={Springer}
}

Repository for self-supervised landmark discovery

Related tags

Overview

self-supervised-landmarks

Requirements

Usage

Reference

Owner

Riddhish Bhalodia

Local Similarity Pattern and Cost Self-Reassembling for Deep Stereo Matching Networks

EmoTag helps you train emotion detection model for Chinese audios

Code for the paper "Adversarial Generator-Encoder Networks"

This is a repo of basic Machine Learning!

[CVPR 2022] Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels

Understanding the Effects of Datasets Characteristics on Offline Reinforcement Learning

This is the official repository for our paper: ''Pruning Self-attentions into Convolutional Layers in Single Path''.

In real-world applications of machine learning, reliable and safe systems must consider measures of performance beyond standard test set accuracy

A benchmark dataset for emulating atmospheric radiative transfer in weather and climate models with machine learning (NeurIPS 2021 Datasets and Benchmarks Track)

Simple Dynamic Batching Inference

CLOOB: Modern Hopfield Networks with InfoLOOB Outperform CLIP

An implementation of the WHATWG URL Standard in JavaScript

Official implementation of "Variable-Rate Deep Image Compression through Spatially-Adaptive Feature Transform", ICCV 2021

The devkit of the nuScenes dataset.

This repository contains all data used for writing a research paper Multiple Object Trackers in OpenCV: A Benchmark, presented in ISIE 2021 conference in Kyoto, Japan.

A hyperparameter optimization framework

Experiments for Operating Systems Lab (ETCS-352)

Python Actor concurrency library

Python Interview Questions

Create and implement a deep learning library from scratch.