Search

While traditional tracking-by-detection methods have some robustness in object tracking, their simple classification of the target and the background cannot model the relative structural relationship between the target and the background. It is the lack of relative structural discriminative information that always causes a tracker drifting away. In order to alleviate the tracking drifting problem, we propose a new visual travel approach based on deep relative metric learning. In this study, we design a deep relative metric learning model with a symmetric and shared-weight deep convolutional neural network. Through such a network, we can explore the relative structural relationship between the target and the background in large-scale image patches. Then, the highest score of relative metric is used to locate the tracking object in the Bayesian tracking framework. The whole tracking algorithm is simple and effective. Experimental results on the tracking benchmark show that the proposed algorithm achieves a better tracking precision rate and success rate than other state-of-the-art tracking methods.

Visual tracking is still a challenging problem. Tracking drift easily takes place when foreground and background appear similar. In this paper, in order to alleviate the tracking drift problem, we propose a novel visual tracking method by combining multi-view randomized trees and subspace update in an ensemble tracking framework. In our proposed framework, an adaptive multi-view randomized trees is firstly introduced to obtain the accurate confidence map of foreground objects and background. It is the first time that the multi-view randomized trees is introduced into ensemble tracking to improve the accuracy. Secondly, a mean shift tracker detects the object location by seeking the best mode on the confidence map. Moreover, with random forest classifier, the application range of mean shift tracker is also extended from the RGB color space to high-dimensional feature space. To prevent the model from accumulated wrong update, an incremental principal component analysis tracker is implemented as an extra supplement for the ensemble framework to keep the discriminative ability of the tracker. Experimental results have demonstrated that the proposed tracking algorithm consistently provides more powerful ability to decrease the tracking drift than other state-of-the-art approaches.