Comparing Plug-and-Play and Unrolled networks - Signaux et Images
Preprints, Working Papers, ... Year : 2024

Comparing Plug-and-Play and Unrolled networks

Abstract

Plug-and-play and unrolled methods are among the most popular and efficient approaches for solving inverse problems in imaging. In this paper, we review their similarities and differences. We adopt a Bayesian framework to explain their statistical properties. This allows us to relate plug-andplay methods to Maximum A Posteriori (MAP) estimators and unrolled networks to Minimum Mean Square Error (MMSE) estimators. We clarify some of their properties including their stability and numerical efficiency. Overall, this paper provides a concise review of their respective advantages and limitations.

Fichier principal
Vignette du fichier
nguyen_weiss_letter_hal.pdf (1.41 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04703008 , version 1 (20-09-2024)

Licence

Public Domain

Identifiers

  • HAL Id : hal-04703008 , version 1

Cite

Minh Hai Nguyen, Pierre Weiss. Comparing Plug-and-Play and Unrolled networks. 2024. ⟨hal-04703008⟩
160 View
135 Download

Share

More