Affiliations: Institut des Systèmes Intelligents et de Robotique, Université Pierre et Marie Curie/CNRS, UMR 7222, Paris, France | IIT – Italian Institute of Technology Robotics, Brain and Cognitive sciences Department, Genova, Italy
Note: [] Corresponding author. Ilaria Renna, Institut des Systèmes Intelligents et de Robotique, Université Pierre et Marie Curie/CNRS, UMR 7222, Paris, France. Emails: ilaria.renna@upmc.fr (I. Renna), ryad.chellali@iit.it (R. Chellali), catherine.achard@upmc.fr (C. Achard).
Abstract: 3D upper body pose estimation is a topic greatly studied by the computer vision society because it is useful in a great number of applications, mainly for human robots interactions including communications with companion robots. However there is a challenging problem: the complexity of classical algorithms that increases exponentially with the dimension of the vectors' state becomes too difficult to handle. To tackle this problem, we propose a new approach that combines several annealing particle filters defined independently for each limb and belief propagation method to add geometrical constraints between individual filters. Experimental results on a real human gestures sequence will show that this combined approach leads to reliable results.
Keywords: Body tracking, particle filter, belief propagation
