Template matching tracking method based on target decomposition
Chunbo Xiu, Hongjie Li · 2016
In order to improve the tracking performance of the target in complex circumstances, a novel template matching tracking method is proposed by decomposing the target into multiple sub-targets with the distinguishing features. The tracked target contains multiple areas which can be decomposed into multiple sub-targets. Each sub-target is tracked separately by the template matching algorithm. The error locations of sub-targets can be corrected according to the relative positions among all the sub-targets. The correct tracking results can be obtained by the error correction as long as some of the sub-targets are located correctly. Experimental results show that the method proposed in this paper can be adaptive to the scaling and occlusion of the target.