Abstract
Recently, the correlation filter (CF)-based methods have achieved great success in the field of object tracking. In most of these methods, the CF utilizes L 2 norm as the regularization, which does not pay attention to the stability and robustness of the feature. However, there may exist some unstable points in the image because the object in the video may have different appearance changes. We propose a tracking method based on a structured robust correlation filter (SRCF), which employs the L 2,1 norm as the regularization. The robust CF can not only retain the accuracy from the regression formulation but also take into account the stability of the image region to improve the robustness of the appearance model. The alternating direction method of multipliers algorithm is used to solve the L 2,1 optimization problem in SRCF. Moreover, the multilayer convolutional features are adopted to further improve the representation accuracy. The proposed method is evaluated in several benchmark datasets, and the results demonstrate that it can achieve comparable performance with respect to the state-of-the-art tracking methods.
Highlights
Visual object tracking is a hot research topic in the domains of computer vision, multimedia, etc
Besides the OTB-2013 dataset in which the structured robust correlation filter (SRCF) tracker has achieved good results, we evaluate its performance in more datasets, including the Tcolor[128] dataset,[46] OTB100 dataset,[47] VOT2016 dataset,[48] and VOT201749 dataset to explore the effect of the settings of the tracker
We further evaluate the performance of SRCF in OTB-100 dataset, which includes 100 different sequences
Summary
Visual object tracking is a hot research topic in the domains of computer vision, multimedia, etc. Some tracking methods based on a correlation filter (CF), which corresponds to the regression formulation, have achieved great success.[18,19,20,21] On one hand, the CF addresses the sparse sampling in binary classification model, which makes full use of the spatial information. Some researchers[28,29,30,31] have proposed some new tracking methods, which utilize both the deep CNN and CF to further improve the tracking performance. We propose a tracking method based on the structured robust correlation filter (SRCF) with L2;1 norm. To address the impact of the unreliable points in the image region with a multichannel feature, we develop a robust CF and formulate tracking as a structured robust regression problem.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.