Saved in:
Bibliographic Details
Main Authors: Modrzyk, Thibaut, Etxebeste, Ane, Bretin, Élie, Maxim, Voichita
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.24156
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • In this paper, we present a novel variational plug-and-play algorithm for Poisson inverse problems. Our approach minimizes an explicit functional which is the sum of a Kullback-Leibler data fidelity term and a regularization term based on a pre-trained neural network. By combining classical likelihood maximization methods with recent advances in gradient-based denoisers, we allow the use of pre-trained Gaussian denoisers without sacrificing convergence guarantees. The algorithm is formulated in the majorization-minimization framework, which guarantees convergence to a stationary point. Numerical experiments confirm state-of-the-art performance in deconvolution and tomography under moderate noise, and demonstrate clear superiority in high-noise conditions, making this method particularly valuable for nuclear medicine applications.