Rigid and Articulated Point Registration with Expectation Conditional Maximization

Radu Horaud 1 Florence Forbes 2 Manuel Yguel 3 Guillaume Dewaele 1, 4 Jian Zhang 5
1 PERCEPTION - Interpretation and Modelling of Images and Videos
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
2 MISTIS - Modelling and Inference of Complex and Structured Stochastic Systems
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann, INPG - Institut National Polytechnique de Grenoble
3 E-MOTION - Geometry and Probability for Motion and Action
Inria Grenoble - Rhône-Alpes, LIG - Laboratoire d'Informatique de Grenoble
4 Phys-ENS - Laboratoire de Physique de l'ENS Lyon
5 EEE - Department of Electrical and Electronic Engineering
Abstract : This paper addresses the issue of matching rigid and articulated shapes through probabilistic point registration. The problem is recast into a missing data framework where unknown correspondences are handled via mixture models. Adopting a maximum likelihood principle, we introduce an innovative EM-like algorithm, namely the Expectation Conditional Maximization for Point Registration (ECMPR) algorithm. The algorithm allows the use of general covariance matrices for the mixture model components and improves over the isotropic covariance case. We analyse in detail the associated consequences in terms of estimation of the registration parameters, and we propose an optimal method for estimating the rotational and translational parameters based on semi-definite positive relaxation. We extend rigid registration to articulated registration. Robustness is ensured by detecting and rejecting outliers through the addition of a uniform component to the Gaussian mixture model at hand. We provide an in-depth analysis of our method and we compare it both theoretically and experimentally with other robust methods for point registration.
Keywords : outlier detection robust statistics hand tracking feature matching convex optimization expectation maximization EM Gaussian mixture models ICP Point registration SDP relaxation articulated object tracking object pose
Type de document :
Article dans une revue
IEEE Transactions on Pattern Analysis and Machine Intelligence, Institute of Electrical and Electronics Engineers, 2011, 33 (3), pp.587-602. 〈10.1109/TPAMI.2010.94〉
Liste complète des métadonnées

Littérature citée [57 références]  Voir  Masquer  Télécharger
https://hal.inria.fr/inria-00590265
Contributeur : Team Perception <>
Soumis le : jeudi 16 juin 2011 - 11:01:18
Dernière modification le : jeudi 11 octobre 2018 - 08:48:02
Document(s) archivé(s) le : samedi 17 septembre 2011 - 02:20:45

Fichier

HoraudForbesYguelDewaeleZhang-...
Fichiers éditeurs autorisés sur une archive ouverte

Identifiants

Collections

INRIA | ENS-LYON | LJK_PS_MISTIS | LJK_PS | LJK_GI | LJK_GI_PERCEPTION | LJK | LIG | UGA | LIG_SIC_E-MOTION | LIG_SRCPR | UNIV-LYON1

Citation

Radu Horaud, Florence Forbes, Manuel Yguel, Guillaume Dewaele, Jian Zhang. Rigid and Articulated Point Registration with Expectation Conditional Maximization. IEEE Transactions on Pattern Analysis and Machine Intelligence, Institute of Electrical and Electronics Engineers, 2011, 33 (3), pp.587-602. 〈10.1109/TPAMI.2010.94〉. 〈inria-00590265〉

Exporter

BibTeX TEI DC DCterms EndNote

Partager

Métriques

Consultations de la notice

709

Téléchargements de fichiers

498

Contact


archive-ouverte@inria.fr