A Super-resolution Framework for High-Accuracy Multiview Reconstruction

Bastian Goldluecke 1 Mathieu Aubry 2, 3, 4, 5 Kalin Kolev 6 Daniel Cremers 7
1 HCI - Heidelberg Collaboratory for Image Processing
2 ENPC - École des Ponts ParisTech
3 IMAGINE [Marne-la-Vallée]
CSTB - Centre Scientifique et Technique du Bâtiment, LIGM - Laboratoire d'Informatique Gaspard-Monge, ENPC - École des Ponts ParisTech
4 WILLOW - Models of visual object recognition and scene understanding
CNRS - Centre National de la Recherche Scientifique : UMR8548, Inria Paris-Rocquencourt, DI-ENS - Département d'informatique de l'École normale supérieure
5 LIGM - Laboratoire d'Informatique Gaspard-Monge
6 Computer Vision Laboratory - ETHZ [Zurich]
7 CVPR - Computer Vision and Pattern Recognition Group [Münster]
Abstract : We present a variational framework to estimate super-resolved texture maps on a 3D geometry model of a surface from multiple images. Given the calibrated images and the reconstructed geometry, the proposed functional is convex in the super-resolution texture. Using a conformal atlas of the surface, we transform the model from the curved geometry to the flat charts and solve it using state-of-the-art and provably convergent primal-dual algorithms. In order to improve image alignment and quality of the texture, we extend the functional to also optimize for a normal displacement map on the surface as well as the camera calibration parameters. Since the sub-problems for displacement and camera parameters are non-convex, we revert to relaxation schemes in order to robustly estimate a minimizer via sequential convex programming. Experimental results confirm that the proposed super-resolution framework allows to recover textured models with significantly higher level-of-detail than the individual input images.
Keywords : Variational methods Super-resolution Camera calibration Multi-view 3D reconstruction Texture reconstruction
Type de document :
Article dans une revue
International Journal of Computer Vision, Springer Verlag, 2014, 106 (2), pp.172-191. 〈10.1007/s11263-013-0654-8〉
Liste complète des métadonnées

Littérature citée [45 références]  Voir  Masquer  Télécharger
https://hal.inria.fr/hal-01057502
Contributeur : Mathieu Aubry <>
Soumis le : vendredi 22 août 2014 - 17:44:20
Dernière modification le : jeudi 5 juillet 2018 - 14:29:03
Document(s) archivé(s) le : jeudi 27 novembre 2014 - 13:56:46

Fichier

GAKC13_ijcv.pdf
Fichiers produits par l'(les) auteur(s)

Identifiants

Collections

INRIA | ENPC | IMAGINE | ENS-PARIS | LIGM_A3SI | LIGM | PARISTECH | UPEC-UPEM | PSL | INRIA-ROCQ | ESIEE-PARIS

Citation

Bastian Goldluecke, Mathieu Aubry, Kalin Kolev, Daniel Cremers. A Super-resolution Framework for High-Accuracy Multiview Reconstruction. International Journal of Computer Vision, Springer Verlag, 2014, 106 (2), pp.172-191. 〈10.1007/s11263-013-0654-8〉. 〈hal-01057502〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

9311

Téléchargements de fichiers

1081

Contact


archive-ouverte@inria.fr