DIRECT

DIRECT is the Deep Image REConstruction Toolkit that implements state-of-the-art inverse problem solvers. It includes the Recurrent Inference Machine, which was part of the winning solution in Facebook & NYUs FastMRI challenge in 2019 and the Calgary-Campinas MRI reconstruction challenge at MIDL 2020.

Installation

See install.md.

Quick Start

See getting_started.md, check out the documentation. In the projects folder examples are given on how to train models on public datasets.

Baselines and trained models.

We intend to provide a set of baseline results and trained models in the DIRECT Model Zoo.

License

DIRECT is released under the Apache 2.0 License.

Citing DIRECT

If you use DIRECT in your own research, or want to refer to baseline results published in the DIRECT Model Zoo, please use the following BiBTeX entry:

@misc{DIRECTTOOLKIT,
  author =       {Jonas Teuwen, Nikita Moriakov, Dimitrios Karkalousos, Matthan Caan},
  title =        {DIRECT},
  howpublished = {\url{https://github.com/directgroup/direct}},
  year =         {2020}
}