PINER: Prior-Informed Implicit Neural Representation Learning for Test-Time Adaptation in Sparse-View CT Reconstruction

Overview
Installation Guide
Instructions for Running Code
License
Citation

1. Overview

This repository provides the PyTorch code for our WACV 2023 papar PINER: Prior-Informed Implicit Neural Representation Learning for Test-Time Adaptation in Sparse-View CT Reconstruction.

by Bowen Song, Liyue Shen, Lei Xing

We propose a two-stage input-adaptation and output-correction framework with implicit neural representation learning for test-time adaptation for sparse-view CT reconstruction with unknown and varying noise levels.

2. Installation Guide

Run Before running this package, users should have Python, PyTorch, and several python packages installed (numpy, skimage, yaml, opencv, odl) .

Package Versions

This code functions with following dependency packages. The versions of software are, specifically:

python: 3.7.4
pytorch: 1.4.1
numpy: 1.19.4
skimage: 0.17.2
yaml: 0.1.7
opencv: 3.4.2
odl: 1.0.0.dev0
astra-toolbox
torch-radon

Package Installment

Users should install all the required packages shown above prior to running the algorithm. Most packages can be installed by running following command in terminal on Linux. To install PyTorch, please refer to their official website. To install ODL, please refer to their official website.

pip install package-name

3. Instructions for Running Code

2D CT Reconstruction Experiment

The experiments of 2D CT image reconstruction use the 2D parallel-beam geometry.

Step 0: Input Adapataion

Get an adapted input based on output analysis train_image_regression_test.py

Step 1: Prior embedding

Represent 2D image by implicit network network.

python train_image_regression.py --config configs/image_regression.yaml

Step 2: Network training

Reconstruct 2D CT image from sparsely sampled projections. With prior embedding: python train_ct_recon2.py --config configs/ctrecon.yaml

Three Step Pipeline

python PINER_pipeline.py

5. Citation

If you find the code are useful, please consider citing the paper and the related paper.

@inproceedings{song2023piner,
  title={PINER: Prior-Informed Implicit Neural Representation Learning for Test-Time Adaptation in Sparse-View CT Reconstruction},
  author={Song, Bowen and Shen, Liyue and Xing, Lei},
  booktitle={Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision},
  pages={1928--1938},
  year={2023}
}

@article{shen2022nerp,
  title={NeRP: implicit neural representation learning with prior embedding for sparsely sampled image reconstruction},
  author={Shen, Liyue and Pauly, John and Xing, Lei},
  journal={IEEE Transactions on Neural Networks and Learning Systems},
  year={2022},
  publisher={IEEE}
}

efzero / PINER