Chapitre D'ouvrage Année : 2010

Statistical gesture models for 3D motion capture from a library of gestures with variants

Résumé

A challenge for 3D motion capture by monocular vision is 3D-2D projection ambiguities that may bring incorrect poses during tracking. In this paper, we propose improving 3D motion capture by learning human gesture models from a library of gestures with variants. This library has been created with virtual human animations. Gestures are described as Gaussian Process Dynamic Models (GPDM) and are used as constraints for motion tracking. Given the raw input poses from the tracker, the gesture model helps to correct ambiguous poses. The benefit of the proposed method is demonstrated with results.

Fichier principal
Vignette du fichier
GW09_LNCS_Horain_Pez_Pelachaud.pdf (532.38 Ko) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence
Loading...

Dates et versions

hal-00473788 , version 1 (16-04-2010)

Licence

Identifiants

  • HAL Id : hal-00473788 , version 1

Citer

Zhenbo Li, Patrick Horain, André-Marie Pez, Catherine Pelachaud. Statistical gesture models for 3D motion capture from a library of gestures with variants. Gesture in Embodied Communication and Human-Computer Interaction : 8th International Gesture Workshop, GW 2009, Bielefeld, Germany, February 25-27, 2009, Springer, 2010, Lecture notes in artificial intelligence. ⟨hal-00473788⟩
455 Consultations
308 Téléchargements

Partager

  • More