Share Email Print
cover

Proceedings Paper

Object tracking algorithm based on contextual visual saliency
Author(s): Bao Fu; XianRong Peng
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

As to object tracking, the local context surrounding of the target could provide much effective information for getting a robust tracker. The spatial-temporal context (STC) learning algorithm proposed recently considers the information of the dense context around the target and has achieved a better performance. However STC only used image intensity as the object appearance model. But this appearance model not enough to deal with complicated tracking scenarios. In this paper, we propose a novel object appearance model learning algorithm. Our approach formulates the spatial-temporal relationships between the object of interest and its local context based on a Bayesian framework, which models the statistical correlation between high-level features (Circular-Multi-Block Local Binary Pattern) from the target and its surrounding regions. The tracking problem is posed by computing a visual saliency map, and obtaining the best target location by maximizing an object location likelihood function. Extensive experimental results on public benchmark databases show that our algorithm outperforms the original STC algorithm and other state-of-the-art tracking algorithms.

Paper Details

Date Published: 27 September 2016
PDF: 6 pages
Proc. SPIE 9684, 8th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optical Test, Measurement Technology, and Equipment, 96842O (27 September 2016); doi: 10.1117/12.2243216
Show Author Affiliations
Bao Fu, Institute of Optics and Electronics (China)
Univ. of Chinese Academy of Sciences (China)
XianRong Peng, Institute of Optics and Electronics (China)


Published in SPIE Proceedings Vol. 9684:
8th International Symposium on Advanced Optical Manufacturing and Testing Technologies: Optical Test, Measurement Technology, and Equipment
Yudong Zhang; Fan Wu; Ming Xu; Sandy To, Editor(s)

© SPIE. Terms of Use
Back to Top