PT Journal
AU Ignasi Rius
   Jordi Gonzalez
   J. Varona
   Xavier Roca
TI Action-specific motion prior for efficient bayesian 3D human body tracking
SO Pattern Recognition
JI PR
PY 2009
BP 2907–2921
VL 42
IS 11
DI 10.1016/j.patcog.2009.02.012
AB In this paper, we aim to reconstruct the 3D motion parameters of a human bodymodel from the known 2D positions of a reduced set of joints in the image plane.Towards this end, an action-specific motion model is trained from a database of realmotion-captured performances. The learnt motion model is used within a particlefiltering framework as a priori knowledge on human motion. First, our dynamicmodel guides the particles according to similar situations previously learnt. Then, the solution space is constrained so only feasible human postures are accepted as valid solutions at each time step. As a result, we are able to track the 3D configuration of the full human body from several cycles of walking motion sequences using only the 2D positions of a very reduced set of joints from lateral or frontal viewpoints.
ER