Journal Article

Deep learning methods for inverse problems

Azimifar, Zohreh, Sabzi, Rasool, Kamyab, Shima, Fieguth, Paul

2022

Overview

In this paper we investigate a variety of deep learning strategies for solving inverse problems. We classify existing deep learning solutions for inverse problems into three categories of Direct Mapping, Data Consistency Optimizer, and Deep Regularizer. We choose a sample of each inverse problem type, so as to compare the robustness of the three categories, and report a statistical analysis of their differences. We perform extensive experiments on the classic problem of linear regression and three well-known inverse problems in computer vision, namely image denoising, 3D human face inverse rendering, and object tracking, in presence of noise and outliers, are selected as representative prototypes for each class of inverse problems. The overall results and the statistical analyses show that the solution categories have a robustness behaviour dependent on the type of inverse problem domain, and specifically dependent on whether or not the problem includes measurement outliers. Based on our experimental results, we conclude by proposing the most robust solution category for each inverse problem class.

• Publisher: PeerJ. Ltd,PeerJ, Inc,PeerJ Inc
• Subject: 3D reconstruction as inverse problem / Artificial Intelligence / Categories / Computer Vision / Deep learning / Deep learning for inverse problems / Image denoising as inverse problem / Inverse problems / Knowledge / Learning strategies / Linear regression as inverse problem / Literature reviews / Machine vision / Methods / Noise / Outliers (statistics) / Performance evaluation / Robustness (mathematics) / Single object tracking as inverse problem / Statistical analysis / Training