Measurement-Consistent Networks via a Deep Implicit Layer for Solving Inverse Problems

Rahul Mourya, João F. C. Mota

Research output: Chapter in Book/Report/Conference proceedingConference contribution

22 Downloads (Pure)

Abstract

End-to-end deep neural networks (DNNs) have become state-of-the-art (SOTA) for solving inverse problems. Despite their outstanding performance, during deployment, such networks are sensitive to minor variations in the training pipeline and often fail to reconstruct small but important details, a feature critical in medical imaging, astronomy, or defence. Such instabilities in DNNs can be explained by the fact that they ignore the forward measurement model during deployment, and thus fail to enforce consistency between their output and the input measurements. To overcome this, we propose a framework that transforms any DNN for inverse problems into a measurement-consistent one. This is done by appending to it an implicit layer (or deep equilibrium network) designed to solve a model-based optimization problem. The implicit layer consists of a shallow learnable network that can be integrated into the end-to-end training. Experiments on single-image super-resolution show that the proposed framework leads to significant improvements in reconstruction quality and robustness over the SOTA DNNs.
Original languageEnglish
Title of host publication2023 IEEE International Conference on Acoustics, Speech and Signal Processing
Publication statusAccepted/In press - 17 Feb 2023
Event48th IEEE International Conference on Acoustics, Speech and Signal Processing 2023 - Rhodes Island, Rhodes Island, Greece
Duration: 4 Jun 202310 Jun 2023
Conference number: 48
https://2023.ieeeicassp.org/
https://2023.ieeeicassp.org

Conference

Conference48th IEEE International Conference on Acoustics, Speech and Signal Processing 2023
Abbreviated titleICASSP 2023
Country/TerritoryGreece
CityRhodes Island
Period4/06/2310/06/23
Internet address

Keywords

  • cs.CV
  • eess.IV

Fingerprint

Dive into the research topics of 'Measurement-Consistent Networks via a Deep Implicit Layer for Solving Inverse Problems'. Together they form a unique fingerprint.

Cite this