Augmented particle samples based optimal convolutional filters for object tracking
- PDF / 3,122,557 Bytes
- 19 Pages / 439.642 x 666.49 pts Page_size
- 67 Downloads / 245 Views
Augmented particle samples based optimal convolutional filters for object tracking Xiaowei An1 · Quanquan Liang2
· Nongliang Sun2
Received: 27 February 2020 / Revised: 4 August 2020 / Accepted: 25 August 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract This paper presents the augmented particle samples based optimal convolutional filters that preserve the appearance model robustness for object tracking in both temporal and spatial levels. In temporal level, augmented particle samples provided by Laplacian group reverse sparse representation exploit the potential geometrical correlation among the different patches that keep the inherent potential distribution which facilitates the update scheme of appearance model between continuous frames in the particle filtering framework. In spatial level, structural information of multi-scale patches extraction can preserve highly stable attributes that significantly improve the object representation robustness in multi-scenarios. Moreover, the optimal convolutional filters that resulted from laplacian score exploits the coherence of high similarity in both positive and negative sets effectively that can guarantee the template update procedures discriminatively. Experimental results demonstrate that the proposed approach achieves better performance on multiple dynamic scenes. Keywords Augmented particle samples · Optimal convolutional filters · Laplacian group reverse sparse representation · Laplacian score · Particle filtering
Quanquan Liang
[email protected] Nongliang Sun
nl [email protected]; [email protected] Xiaowei An [email protected] 1
College of Electrical Engineering and Automation, Shandong University of Science and Technology, Qingdao, 266590, Shandong Province People’s Republic of China
2
College of Electronics and Information Engineering, Shandong University of Science and Technology, Qingdao, 266590, Shandong Province, People’s Republic of China
Multimedia Tools and Applications
1 Introduction Although numbers of algorithms have been proposed for massive challenging problems in the object tracking, such as illumination, pose variation, environmental cluttering, and object occlusion. Accuracy and robustness problems are still two challenging obstacles for simple and useful processing techniques in this realm [34, 40]. In general, observation model and motion model are two most crucial parts in tracking process. It is very significant to find the appropriate regions which own the most similar target appearance description according to the generative tracking templates, or train efficient classifiers for identifying foreground and background information. Incorporating multiple machine learning techniques, many approaches have been devoted into boosting the classifiers discriminatively which give a decision of the boundary between target and background. In [8], researches applied the robust feature selection with boosting cascade weak classifiers for more strong classifiers. In [14], the P -N learnin
Data Loading...
