English

How to do Physics-based Learning

Image and Video Processing 2020-05-29 v2 Computer Vision and Pattern Recognition Signal Processing

Abstract

The goal of this tutorial is to explain step-by-step how to implement physics-based learning for the rapid prototyping of a computational imaging system. We provide a basic overview of physics-based learning, the construction of a physics-based network, and its reduction to practice. Specifically, we advocate exploiting the auto-differentiation functionality twice, once to build a physics-based network and again to perform physics-based learning. Thus, the user need only implement the forward model process for their system, speeding up prototyping time. We provide an open-source Pytorch implementation of a physics-based network and training procedure for a generic sparse recovery problem

Cite

@article{arxiv.2005.13531,
  title  = {How to do Physics-based Learning},
  author = {Michael Kellman and Michael Lustig and Laura Waller},
  journal= {arXiv preprint arXiv:2005.13531},
  year   = {2020}
}

Comments

3 pages, 2 figures, linked repository https://github.com/kellman/physics_based_learning