Fast nonparametric belief propagation for real-time stereo articulated body tracking

This article proposes a statistical approach for fast articulated 3D body tracking, similar to the loose-limbed model, but using the factor graph representation and a fast estimation algorithm. A fast Nonparametric Belief Propagation on factor graphs is used to estimate the current marginal for each limb. All belief propagation messages are represented as sums of weighted samples. The resulting algorithm corresponds to a set of particle filters, one for each limb, where an extra step recomputes the weight of each sample by taking into account the links between limbs. Applied to upper body tracking with stereo and colour images, the resulting algorithm estimates the body pose in quasi real-time (10 Hz). Results on sequences illustrate the effectiveness of this approach.