Learning discriminative update adaptive spatial-temporal regularized correlation filter for RGB-T tracking

• Published in: Journal of visual communication and image representation Vol. 72; p. 102881
• Main Authors: Feng, Mingzheng, Song, Kechen, Wang, Yanyan, Liu, Jie, Yan, Yunhui
• Format: Journal Article
• Language: English
• Published: Elsevier Inc 01.10.2020
• Subjects: Adaptive spatial-temporal regularization; ADMM; Correlation filters; Model updating; Object tracking; Adaptive spatial-temporal regularization; Object tracking; ADMM; Correlation filters; Model updating
• ISSN: 1047-3203; 1095-9076
• DOI: 10.1016/j.jvcir.2020.102881

•An adaptive spatial-temporal regularized correlation filter model is proposed to achieve robust tracking.•A relative peak discriminative method for model updating is proposed to avoid tracker drifting.•An adaptive weighting ensemble scheme is utilized to obtain the complementary benefits of the RGB and thermal modes.•An effective ADMM algorithm is designed to optimize the RGB-T correlation filter model.•Experimental results demonstrate the effectiveness of the proposed tracker against other state-of-the-art RGB-T trackers. The RGB-T trackers based on correlation filter framework have been extensively investigated for that they can track targets more accurately in most complex scenes. However, the performance of these trackers is limited when facing some specific challenging scenarios, such as occlusion and background clutter. For different tracking targets, most of these trackers utilize fixed regularization constraint to build the filter model, which is obviously unreasonable to effectively present the appearance changes and characteristics of a specific target. In addition, they adopt a simple model update mechanism based on linear interpolation, which can easily lead to model degradation in challenging scenarios, resulting in tracker drift. To solve the above problems, we propose a novel adaptive spatial-temporal regularized correlation filter model to learn an appropriate regularization for achieving robust tracking and a relative peak discriminative method for model updating to avoid the model degradation. Besides, to make better integrate the unique advantages of the two modes and adapt the changing appearance of the target, an adaptive weighting ensemble scheme and a multi-scale search mechanism are adopted, respectively. To optimize the proposed model, we designed an efficient ADMM algorithm, which greatly improved the efficiency. Extensive experiments have been carried out on two available datasets, RGBT234 and RGBT210, and the experimental results indicate that the tracker proposed by us performs favorably in both accuracy and robustness against the state-of-the-art RGB-T trackers.