Tracking of Rapid Motions Using Particle Filtering Based on Grid Resampling

These complex real environmental videos usually contain rapid motions. The popular traditional particle filtering algorithms are unable to tackle this issue. In order to address it, we propose a novel tracking algorithm for rapid motions using particle filtering based on grid resampling, the resampling can break smoothness constraints of motion model, can robustly capture the effect of abnormal motion. A series of experimental results show the good performance of the proposed method.