Hierarchical Association Model Based Target Tracking

This paper proposes a hierarchical association multi-target tracking trajectory generation method. The AdBoost method combined with online discriminant analysis apparent model is first utilized to achieve initial tracking trajectories; then, the Hungarian method is utilized to optimize fragmented and discontinuous trajectories to achieve accurate trajectories fragments; finally, the intelligent extrapolation based on the energy minimization utilized to achieve the final smoother and continuous trajectories. Experimental results on PETS 2009/2010 benchmark demonstrate the effectiveness and efficiency of the proposed scheme.