Robust Histogram-Based Object Tracking in Image Sequences

Object tracking is an important aspect of computer vision which has attracted considerable attention recently. In this paper, we first introduce our framework for object tracking. Then, a new mathematical model for tracking performance analysis is presented. In addition, five matrix filters are designed and adopted to realize robust and efficient object tracking. Finally, an approach for combining histograms with different feature spaces into one histogram is given. According to our extensive experimental results, we found that our approaches greatly improve the visual tracking results.