Discrete Mumford-Shah on graph for mixing matrix estimation - INRIA - Institut National de Recherche en Informatique et en Automatique Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2018

Discrete Mumford-Shah on graph for mixing matrix estimation

Résumé

The discrete Mumford-Shah formalism has been introduced for the image denoising problem, allowing to capture both smooth behavior inside an object and sharp transitions on the boundary. In the present work, we propose first to extend this formalism to graphs and to the problem of mixing matrix estimation. New algorithmic schemes with convergence guarantees relying on proximal alternating minimization strategies are derived and their efficiency (good estimation and robustness to initialization) are evaluated on simulated data, in the context of vote transfer matrix estimation.
Fichier principal
Vignette du fichier
double.pdf (978.02 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01948570 , version 1 (07-12-2018)
hal-01948570 , version 2 (27-05-2019)

Identifiants

  • HAL Id : hal-01948570 , version 1

Citer

Yacouba Kaloga, Marion Foare, Nelly Pustelnik, Pablo Jensen. Discrete Mumford-Shah on graph for mixing matrix estimation. 2018. ⟨hal-01948570v1⟩
285 Consultations
375 Téléchargements

Partager

Gmail Facebook X LinkedIn More