Object tracking using KLT aided mean-shift object tracker (ICCAS 2014)

In this paper, we present a new object tracking algorithm which integrates color-based mean-shift and feature-based optical flow methods. To utilize two approaches in the complimentary manner, we iteratively compute the mean-shift vector based on color histograms and tracked features by KLT. In the experiments, we show the improved performance for partial occlusion and severe appearance changes in the representative benchmark sequences.