112 search results - page 8 / 23 » Temporal motion models for monocular and multiview 3D human ... |
ICVGIP
2004
13 years 9 months ago
2004
We present a method to determine the 3D spatial locations of joints of a human body from a monocular video sequence of a Bharatanatyam dance. The proposed method uses domain speci...
CVIU
2006
13 years 7 months ago
2006
This paper proposes a clustered exemplar-based model for performing viewpoint invariant tracking of the 3D motion of a human subject from a single camera. Each exemplar is associa...
ICPR
2008
IEEE
14 years 1 months ago
2008
IEEE
We propose a Dynamic Bayesian Network (DBN) model for upper body tracking. We first construct a Bayesian Network (BN) to represent the human upper body structure and then incorpo...
GW
2009
Springer
13 years 5 months ago
2009
Springer
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 ...
CVPR
2010
IEEE
14 years 4 months ago
2010
IEEE
While the problem of tracking 3D human motion has been widely studied, most approaches have assumed that the person is isolated and not interacting with the environment. Environme...