Invertible Learned Primal-Dual

Abstract

We propose invertible Learned Primal-Dual as a method for tomographic image reconstruction. This is a learned iterative method based on the Learned Primal-Dual neural network architecture, which incorporates ideas from invertible neural networks. The invertibility significantly reduces the GPU memory footprint of the Learned Primal-Dual architecture, thus making it applicable to 3D tomographic reconstruction as demonstrated in the experiments.

Cite

Text

Rudzusika et al. "Invertible Learned Primal-Dual." NeurIPS 2021 Workshops: Deep_Inverse, 2021.

Markdown

[Rudzusika et al. "Invertible Learned Primal-Dual." NeurIPS 2021 Workshops: Deep_Inverse, 2021.](https://mlanthology.org/neuripsw/2021/rudzusika2021neuripsw-invertible/)

BibTeX

@inproceedings{rudzusika2021neuripsw-invertible,
  title     = {{Invertible Learned Primal-Dual}},
  author    = {Rudzusika, Jevgenija and Bajic, Buda and Öktem, Ozan and Schönlieb, Carola-Bibiane and Etmann, Christian},
  booktitle = {NeurIPS 2021 Workshops: Deep_Inverse},
  year      = {2021},
  url       = {https://mlanthology.org/neuripsw/2021/rudzusika2021neuripsw-invertible/}
}