A mathematical framework unifying various Shape from Shading approaches

Emmanuel Prados 1 Olivier Faugeras 1 Fabio Camilli 2
1 ODYSSEE - Computer and biological vision
DI-ENS - Département d'informatique de l'École normale supérieure, CRISAM - Inria Sophia Antipolis - Méditerranée , ENS Paris - École normale supérieure - Paris, Inria Paris-Rocquencourt, ENPC - École des Ponts ParisTech
Abstract : By slightly modifying the notion of singular viscosity solutions [Ishii-Ramaswamy:95,Camilli-Siconolfi:99,Camilli:01,Camilli-Siconolfi:02] we define a new mathematical framework allowing to unify the various theoretical results proposed in the Shape from shading literature. We demonstrate the existence and the uniqueness of the new solution for a class of Hamilton-Jacobi equations including the classical Shape-From-Shading equations [Prados-Faugeras:03], in a bounded locally Lipschitz domain. Some stability results are proved. Finally, we propose a provably convergent numerical method for approximating the solution and we demonstrate its relevance and its efficiency by numerical experiments on real images.
Type de document :
Communication dans un congrès
Mathematics and Image Analysis (MIA '04), Sep 2004, Paris, France. 2004
Liste complète des métadonnées

https://hal.inria.fr/inria-00590178
Contributeur : Team Perception <>
Soumis le : mardi 3 mai 2011 - 09:29:48
Dernière modification le : jeudi 11 janvier 2018 - 16:23:57

Identifiants

  • HAL Id : inria-00590178, version 1

Collections

Citation

Emmanuel Prados, Olivier Faugeras, Fabio Camilli. A mathematical framework unifying various Shape from Shading approaches. Mathematics and Image Analysis (MIA '04), Sep 2004, Paris, France. 2004. 〈inria-00590178〉

Partager

Métriques

Consultations de la notice

235