Based On Weighted Correlation And Adaptive Context Visual Tracking

Visual tracking is one of the most fundamental problems in computer vision with various applications such as security and surveillance,human computer interaction and vehicle navigation.Despite the significant progress,it remains a challenging task for a tracker to distinguish a target from the background when the target is occluded.In this paper,we propose two new tracking methods: depth-weighted correlation method and context adaptive learning method,to handle heavy occlusion.The proposed depth-weighted correlation method uses depth cues as the weights of candidate objects and applies the framework of spatio-temporal context(STC).We also propose a scale update scheme for depth-weighted correlation method,so as to obtain an appropriate scale for the target.The proposed context adaptive learning method exploits the intrinsic relationship among target regions and background regions for distinguishing the distracting patches.We also design a biologically inspired approach to further improve the performance of visual tracking.Experiments on challenging benchmark image sequences demonstrate that the proposed algorithms perform favorably against several state-of-the-art methods.