High Confidence Updating Strategy on Staple Trackers

In the visual tracking, correlation filtering (CF) based on tracking algorithms have shown favorable performance in recent years, and have the impressive performance on benchmark datasets. However, the tracking model has limited information about their context and can easily drift in cases of fast motion, occlusion or background clutter, and the trackers update tracking models at each frame without considering whether the detection is accurate or not. In this paper, we present an improved strategy that is adding more background context and changing the tracker model updating strategy. Experimental results show that the performance of the model has been improved effectively.