Bayesian 3D Human Motion Capture Using Factored Particle Filtering - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2010

Bayesian 3D Human Motion Capture Using Factored Particle Filtering

Abdallah Dib
  • Fonction : Auteur
  • PersonId : 885089
Cédric Rose
  • Fonction : Auteur
  • PersonId : 830760

Résumé

We present a markerless human motion capture system that estimates the 3D positions of the body joints over time. The system uses a dynamic bayesian network and a factored particle filtering algorithm. In this paper we evaluate the impact of using different observation functions for the bayesian state estimation: chamfer distance, a pixel intersection and finally a pseudo-observation of the subject direction calculated from the previous output of the system. We also compare two methods for the factored generation of the particles. The first one uses a deterministic interval exploration strategy whereas the second one is based on an adaptive diffusion. The capacity of the system to recover after occlusion by obstacles was tested on simulated movements in a virtual scene.
Fichier non déposé

Dates et versions

inria-00546925 , version 1 (15-12-2010)

Identifiants

  • HAL Id : inria-00546925 , version 1

Citer

Abdallah Dib, Cédric Rose, François Charpillet. Bayesian 3D Human Motion Capture Using Factored Particle Filtering. 22th International Conference on Tools with Artificial Intelligence - ICTAI 2010, IEEE, Oct 2010, Arras, France. ⟨inria-00546925⟩
83 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More