Visual Hand Tracking Using Nonparametric Belief Propagation

Citations Over TimeTop 10% of 2005 papers

Abstract

This paper develops probabilistic methods for visual tracking of a three-dimensional geometric hand model from monocular image sequences. We consider a redundant representation in which each model component is described by its position and orientation in the world coordinate frame. A prior model is then defined which enforces the kinematic constraints implied by the model's joints. We show that this prior has a local structure, and is in fact a pairwise Markov random field. Furthermore, our redundant representation allows color and edge-based likelihood measures, such as the Chamfer distance, to be similarly decomposed in cases where there is no self-occlusion. Given this graphical model of hand kinematics, we may track the hand's motion using the recently proposed nonparametric belief propagation (NBP) algorithm. Like particle filters, NBP approximates the posterior distribution over hand configurations as a collection of samples. However, NBP uses the graphical structure to greatly reduce the dimensionality of these distributions, providing improved robustness. Several methods are used to improve NBP's computational efficiency, including a novel KD-tree based method for fast Chamfer distance evaluation. We provide simulations showing that NBP may be used to refine inaccurate model initializations, as well as track hand motion through extended image sequences.

Related Papers

• → Stereo matching using belief propagation(2003)1,275 cited
• → Comparison of graph cuts with belief propagation for stereo, using identical MRF parameters(2003)460 cited
• → Stereo Matching Using Belief Propagation(2002)308 cited
• → Applying Sum and Max Product Algorithms of Belief Propagation to 3D Shape Matching and Registration(2009)
• A Study on Fast Stereo Matching Algorithm using Belief Propagation in Multi-resolution Domain(2008)