Increasing multi-trackers robustness with a segmentation algorithm
Marta Marrón-Romera, Miguel Ángel Sotelo, J.C. García · 2006
Abstract:- Different solutions have been proposed for multiple objects tracking based on probabilistic algorithms. In this paper, the authors propose the use of an only particle filter to track a variable number of objects. The robustness and adaptability of the estimator are increased by the use of a clustering algorithm. The measurements used in the tracking process are extracted from a stereovision system, and thus, the 3D position of the tracked objects is obtained at each time step. Tracking results are presented at the end of the paper, showing the reliability of the proposals.