TY - JOUR AU - Daniel Ponsa AU - Antonio Lopez PY - 2009// TI - Variance reduction techniques in particle-based visual contour Tracking T2 - PR JO - Pattern Recognition SP - 2372–2391 VL - 42 IS - 11 KW - Contour tracking KW - Active shape models KW - Kalman filter KW - Particle filter KW - Importance sampling KW - Unscented particle filter KW - Rao-Blackwellization KW - Partitioned sampling N2 - This paper presents a comparative study of three different strategies to improve the performance of particle filters, in the context of visual contour tracking: the unscented particle filter, the Rao-Blackwellized particle filter, and the partitioned sampling technique. The tracking problem analyzed is the joint estimation of the global and local transformation of the outline of a given target, represented following the active shape model approach. The main contributions of the paper are the novel adaptations of the considered techniques on this generic problem, and the quantitative assessment of their performance in extensive experimental work done. L1 - http://refbase.cvc.uab.es/files/PoL2009.pdf UR - http://dx.doi.org/10.1016/j.patcog.2009.04.007 N1 - ADAS ID - Daniel Ponsa2009 ER -