Adversarial Regularizers in Inverse Problems

TL;DR

This paper introduces a neural network-based regularization framework for inverse problems that can be trained with unsupervised data, improving denoising and CT reconstruction.

Contribution

It presents a novel data-driven regularizer for inverse problems that operates with unsupervised training data, bridging model-based and learning-based methods.

Findings

Effective denoising on BSDS dataset

Improved CT reconstruction on LIDC dataset

Framework works with unsupervised training data

Abstract

Inverse Problems in medical imaging and computer vision are traditionally solved using purely model-based methods. Among those variational regularization models are one of the most popular approaches. We propose a new framework for applying data-driven approaches to inverse problems, using a neural network as a regularization functional. The network learns to discriminate between the distribution of ground truth images and the distribution of unregularized reconstructions. Once trained, the network is applied to the inverse problem by solving the corresponding variational problem. Unlike other data-based approaches for inverse problems, the algorithm can be applied even if only unsupervised training data is available. Experiments demonstrate the potential of the framework for denoising on the BSDS dataset and for computed tomography reconstruction on the LIDC dataset.