Hierarchical shape-based surface reconstruction for dense multi-view stereo

Patrick Labatut 1, 2, 3 Jean-Philippe Pons 1, 2, 3 Renaud Keriven 1, 2
2 IMAGINE [Marne-la-Vallée]
LIGM - Laboratoire d'Informatique Gaspard-Monge, CSTB - Centre Scientifique et Technique du Bâtiment, ENPC - École des Ponts ParisTech
Abstract : The recent widespread availability of urban imagery has lead to a growing demand for automatic modeling from multiple images. However, modern image-based modeling research has focused either on highly detailed reconstructions of mostly small objects or on human-assisted simplified modeling. This paper presents a novel algorithm which automatically outputs a simplified, segmented model of a scene from a set of calibrated input images, capturing its essential geometric features. Our approach combines three successive steps. First, a dense point cloud is created from sparse depth maps computed from the input images. Then, shapes are robustly extracted from this set of points. Finally, a compact model of the scene is built from a spatial subdivision induced by these structures: this model is a global minimum of an energy accounting for the visibility of the final surface. The effectiveness of our method is demonstrated through several results on both synthetic and real data sets, illustrating the various benefits of our algorithm, its robustness and its relevance for architectural scenes.
Type de document :
Communication dans un congrès
Liste complète des métadonnées

https://hal-enpc.archives-ouvertes.fr/hal-00834926
Contributeur : Pascal Monasse <>
Soumis le : mardi 18 juin 2013 - 11:56:08
Dernière modification le : jeudi 5 juillet 2018 - 14:25:05
Document(s) archivé(s) le : jeudi 19 septembre 2013 - 04:07:56

Fichier

3DIM09b.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

  • HAL Id : hal-00834926, version 1

Citation

Patrick Labatut, Jean-Philippe Pons, Renaud Keriven. Hierarchical shape-based surface reconstruction for dense multi-view stereo. ICCV, Oct 2009, Kyoto, Japan. pp.1598-1605. ⟨hal-00834926⟩

Partager

Métriques

Consultations de la notice

516

Téléchargements de fichiers

504