Improved hierarchical association model based mult-target tracking
Xiangxiang Li, Songhao Zhu, Lingling Chen, Zhe Shi · 2015
To deal with the issue of multi-target tracking, this paper proposes a hierarchical correlation multi-target tracking trajectory generation method. On the basis of target detection and initial trajectory, the AdBoost algorithm combined with online discriminant analysis apparent model is first utilized to achieve initial tracking trajectories; then, the Hungarian algorithm is here utilized to optimize fragmented and discontinuous tracking trajectories to achieve stable and accurate trajectories fragments; finally, the intelligent extrapolation based on energy minimization here utilized to achieve the final smoother and longer tracking trajectories. Experimental results on PETS 2009/2010 benchmark and TUD-Stadtmitte video database demonstrate the effectiveness and efficiency of the proposed scheme.