Compressive Tracking via Weighted Classification Boosted by Feature Selection

The drifting problem of object tracking is efficiently alleviated in our research. In this paper, an advanced compressive tracking algorithm based on a weighted classifier boosted by feature selection is proposed. The compressed features with high discrimination are selected from the target information of previous and current frames by a discrimination evaluating strategy. These discriminating features are used to train a weighted classifier, which is composed of two sub-classifiers based on previous and current samples bags. Finally, the weighted classifier is used to tell the target object from the background. Experimental results show that the performance in terms of accuracy and robustness hugely improves in tracking via the proposed classification method.