SHI, Hui; TRAONMILIN, Yann; AUJOL, Jean-François

Métadonnées

Afficher la notice complète

Licence d’utilisation du document

SHI, Hui
Institut de Mathématiques de Bordeaux [IMB]

TRAONMILIN, Yann

AUJOL, Jean-François

Langue

Article de revue

Ce document a été publié dans

SIAM Journal on Imaging Sciences. 2022

Society for Industrial and Applied Mathematics

Date de soutenance

2022

Résumé en anglais

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.< Réduire

Mots clés en anglais

Image denoising

Compressive learning

Sketching

Optimization

URI

https://oskar-bordeaux.fr/handle/20.500.12278/191170

Project ANR

Régularisation performante de problèmes inverses en grande dimension pour le traitement de données - ANR-20-CE40-0001

Origine

Importé de hal

Unités de recherche

Institut de Mathématiques de Bordeaux (IMB) - UMR 5251