Covariance matching can track target robustly by fusing multiple features.Unfortunately,covariance tracking is computation demanding for a large target,making it very challenging to track the target in real-time.In order to overcome this shortcoming,a new method based on random sampling is proposed in this paper.We boost the efficiency of covariance tracking based on random sampling which can make the computing time independence of the target's size.Experiment results show that the steady object tracking ca...