We give an overview of algorithms for compressed sensing that are based on neural networks. We experimentally show (using simulated data) that methods using convolutional neural networks are faster than traditional algorithms based on convex optimization, while giving similar results. We then investigate the possible real world usage of this methods for the compression of data acquired by a Time-of-Flight camera.

This presentation is part of Minisymposium “MS65 - Machine learning techniques for image reconstruction (2 parts)”
organized by: Markus Haltmeier (University Innsbruck) , Linh Nguyen (University of Idaho) .

Authors:

Stephan Antholzer (University of Innsbruck)

Keywords:

deep learning, image compression, image reconstruction, machine learning