Skip to content

This code accompanies "Differentiable probabilistic models of scientific imaging with the Fourier slice theorem", UAI 2019

License

Notifications You must be signed in to change notification settings

KarenUllrich/Pytorch-Backprojection

Repository files navigation

Differentiable probabilistic models of scientific imaging with the Fourier slice theorem

Pytorch implementation of differntiable orthogonal integral projection (backprojection) operator in Fourier space, based on our paper:

Differentiable probabilistic models of scientific imaging with the Fourier slice theorem (UAI 2019) Karen Ullrich, Rianne van den Berg, Marcus A. Brubaker, David Fleet, Max Welling

Requirements

The requirements for the conda environment in which we have tested this code are started in requirements.txt. The main dependencies are

  • python 3.6
  • pytorch 1.1.0

A suitable conda environment may be installed via conda env create -f requirements.yml And used by source activate backprojection

Usage

We provide a light introduction to scientific imaging in the jupyter-notebook cryo-tutorial.ipynb. Specifically the generative model of scientific imaging observation_model.py might prove useful for any application that involves orthogonal integral projection.

Maintenance

Please be warned that this repository is not going to be maintained regularly.

Cite

Please cite our paper if you use this code in your own work:

@inproceedings{ullrich2019backprojection,
  title={Differentiable probabilistic models of scientific imaging with the Fourier slice theorem},
  author={Karen Ullrich, Rianne van den Berg,  Marcus A. Brubaker, David Fleet, Max Welling},
  booktitle={proceedings of the Conference on Uncertainty in Artificial Intelligence (UAI)},
  year={2010}
}

About

This code accompanies "Differentiable probabilistic models of scientific imaging with the Fourier slice theorem", UAI 2019

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published