Regularizing the Deep Image Prior with a Learned Denoiser for Linear Inverse Problems - INRIA - Institut National de Recherche en Informatique et en Automatique Accéder directement au contenu
Communication Dans Un Congrès Année : 2021

Regularizing the Deep Image Prior with a Learned Denoiser for Linear Inverse Problems

Résumé

We propose an optimization method coupling a learned denoiser with the untrained generative model, called deep image prior (DIP) in the framework of the Alternating Direction Method of Multipliers (ADMM) method. We also study different regularizers of DIP optimization, for inverse problems in imaging, focusing in particular on denoising and super-resolution. The goal is to make the best of the untrained DIP and of a generic regularizer learned in a supervised manner from a large collection of images. When placed in the ADMM framework, the denoiser is used as a proximal operator and can be learned independently of the considered inverse problem. We show the benefits of the proposed method, in comparison with other regularized DIP methods, for two linear inverse problems, i.e., denoising and super-resolution.
Fichier principal
Vignette du fichier
MMSP-Final.pdf (14.56 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03310533 , version 1 (30-07-2021)

Identifiants

  • HAL Id : hal-03310533 , version 1

Citer

Rita Fermanian, Mikael Le Pendu, Christine Guillemot. Regularizing the Deep Image Prior with a Learned Denoiser for Linear Inverse Problems. MMSP 2021 - IEEE 23rd International Workshop on Multimedia Siganl Processing, Oct 2021, Tampere, Finland. pp.1-6. ⟨hal-03310533⟩
190 Consultations
97 Téléchargements

Partager

Gmail Facebook X LinkedIn More