In this paper, we propose to better estimate high-dimensional distributions by exploiting conditional independences within the Particle Filter (PF) framework. We rst exploit Dynamic Bayesian Networks to determine conditionally independent subspaces of the state space, which allows us to independently perform propagations and corrections over smaller spaces. Second, we propose a swapping process to transform the weighted particle set provided by the update step of PF into a \new particle set" better focusing on high peaks of the posterior distribution. This new methodology, called Swapping-Based Partitioned Sampling, is successfully tested and validated for articulated object tracking