Model Adaptation for Inverse Problems in Imaging

TL;DR

This paper introduces two novel model adaptation methods that enable neural networks to maintain high performance in inverse imaging problems despite changes in the measurement process, without needing additional labeled data.

Contribution

The paper proposes two new procedures for adapting neural networks to changes in the forward model in inverse problems, without requiring extra labeled data.

Findings

Effective in deblurring, super-resolution, and MRI reconstruction

Achieves empirical success across various inverse imaging tasks

Maintains performance despite forward model variations

Abstract

Deep neural networks have been applied successfully to a wide variety of inverse problems arising in computational imaging. These networks are typically trained using a forward model that describes the measurement process to be inverted, which is often incorporated directly into the network itself. However, these approaches are sensitive to changes in the forward model: if at test time the forward model varies (even slightly) from the one the network was trained for, the reconstruction performance can degrade substantially. Given a network trained to solve an initial inverse problem with a known forward model, we propose two novel procedures that adapt the network to a change in the forward model, even without full knowledge of the change. Our approaches do not require access to more labeled data (i.e., ground truth images). We show these simple model adaptation approaches achieve empirical success in a variety of inverse problems, including deblurring, super-resolution, and undersampled image reconstruction in magnetic resonance imaging.