Large displacement object tracking via annealed kernel correlation filter

The Kernelized Correlation Filters (KCF) tracker has caused the extensive concern in recent years because of the high efficiency. Numberous improvements have been made later. However, due to the large displacement motion between the consecutive image frames, these methods cannot track object well. To better cope the problem, we proposed a new simulated annealed KCF (SAKCF) tracker. Take advantage of SA, the proposed method can search global optimal object candidates so as to cover motion space. Meanwhile, In light of the Confidence map, the method provides the unified framework to track smooth or abrupt motion simultaneously. Extensive experiments have demonstrated the effectiveness of our proposed trackers.