DocumentCode :
3317935
Title :
Semi-supervised visual object tracking by label propagation
Author :
Huang, Junheng ; Zhang, Weigang ; Quan, Guangri ; Zhu, Dongjie
Author_Institution :
Sch. of Comput. Sci. & Technol., Harbin Inst. of Technol. at Weihai, Weihai, China
fYear :
2009
fDate :
8-11 Aug. 2009
Firstpage :
560
Lastpage :
564
Abstract :
Recently, object tracking is viewed as a foreground/background two-class classification problem. In this paper, we propose a non-parameter approach to model the observation model for tracking via graph, which is a semi-supervised approach. More specially, the topology structure of graph is carefully designed to reflect the properties of the sample´s distribution during tracking. In predication, the confidence of sample´s label is propagation via random walk with restart (RWR), which can utilize labeled or unlabeled samples easily. The primary advantage of our algorithm is that it keeps the appearance of object in graph model, which can easily model the multi-modal of object appearance. Experimental results demonstrate that, compared with two state of the art methods, the proposed tracking algorithm is more effective, especially in dynamically changing and clutter scenes.
Keywords :
computer vision; graph theory; object detection; foreground-background two-class classification problem; graph model; label propagation; nonparameter approach; semisupervised visual object tracking; topology structure; Computer science; Inference algorithms; Labeling; Layout; Legged locomotion; Principal component analysis; Semisupervised learning; State estimation; Surveillance; Topology;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Computer Science and Information Technology, 2009. ICCSIT 2009. 2nd IEEE International Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-4519-6
Electronic_ISBN :
978-1-4244-4520-2
Type :
conf
DOI :
10.1109/ICCSIT.2009.5234885
Filename :
5234885
Link To Document :
بازگشت