Compressive learning for patch-based image denoising - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue SIAM Journal on Imaging Sciences Année : 2022

Compressive learning for patch-based image denoising

Hui Shi
  • Fonction : Auteur
  • PersonId : 1089606
Yann Traonmilin
  • Fonction : Auteur
  • PersonId : 1077290
Jean-François Aujol

Résumé

The Expected Patch Log-Likelihood algorithm (EPLL) and its extensions have shown good performances for image denoising. The prior model used by EPLL is usually a Gaussian Mixture Model (GMM) estimated from a database of image patches. Classical mixture model estimation methods face computational issues as the high dimensionality of the problem requires training on large datasets. In this work, we adapt a compressive statistical learning framework to carry out the GMM estimation. With this method, called sketching, we estimate models from a compressive representation (the sketch) of the training patches. The cost of estimating the prior from the sketch no longer depends on the number of items in the original large database. To accelerate further the estimation, we add another dimension reduction technique (low-rank modeling of the covariance matrices) to the compressing learning framework. To demonstrate the advantages of our method, we test it on real large-scale data. We show that we can produce denoising performances similar to performances obtained with models estimated from the original training database using GMM priors learned from the sketch with improved execution times.
Fichier principal
Vignette du fichier
ex_article.pdf (6.34 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03429102 , version 1 (18-11-2021)
hal-03429102 , version 2 (14-02-2022)
hal-03429102 , version 3 (21-04-2022)

Identifiants

  • HAL Id : hal-03429102 , version 3

Citer

Hui Shi, Yann Traonmilin, Jean-François Aujol. Compressive learning for patch-based image denoising. SIAM Journal on Imaging Sciences, In press. ⟨hal-03429102v3⟩

Collections

CNRS IMB ANR
232 Consultations
112 Téléchargements

Partager

Gmail Facebook X LinkedIn More