Exploiting Spatial Structure from Parts for Adaptive Kernelized Correlation Filter Tracker

doi:10.1109/LSP.2016.2545705

Exploiting Spatial Structure from Parts for Adaptive Kernelized Correlation Filter Tracker

2016-05-01

22

PRE

AI

DOI：10.1109/LSP.2016.2545705

Original

Original request for help

Save

Abstract

En 中文

Decomposing target into several parts may improve the capability of tracking algorithm to deal with appearance variations such as occlusion and deformation. In this letter, we propose a part-based appearance model by exploiting spatial structure from parts. The model minimizes appearance and deformation cost simultaneously to predict the new position of object. Then, the optimization problem is divided into two parts. Kernelized correlation filter (KCF) is used for tracking the appearance of parts separately to speed up the proposed tracker. Meanwhile, the deformation cost is minimized by structural learning schema, which can reduce the label noise that caused by inaccuracy bounding box. Finally, minimum spanning tree and dynamic programming are employed to combine the score map of the appearance and deformation of parts, and to detect best new position of target. Experimental results on several challenge sequences show the efficiency and effectiveness of the proposed tracking algorithm.

Keywords:

Correlation filter

objects tracking

part-based model

spatial structure

Exploiting Spatial Structure from Parts for Adaptive Kernelized Correlation Filter Tracker

Abstract

Journal

Organization