DocumentCode :
593140
Title :
A Multi-template Decomposed Tracking Method Based on Specific Object of Multi-viewpoint Video and Route Scheme
Author :
Jun Wang ; Jinyong Wu ; Yike Wang ; Zhuo Gong ; Dejian Liu
Author_Institution :
R&D Dept., AnKe Smart City Technol.(PRC)Co., Ltd., Shenzhen, China
fYear :
2012
fDate :
6-8 Nov. 2012
Firstpage :
122
Lastpage :
126
Abstract :
According to the need of specific object´s tracking of multi-viewpoint video, a specific object´s tracking method of multi-viewpoint video based on multi-template decomposed tracking and route scheme has been put forward in this paper. First, the route may be schemed by gridding search in this method. Then the object area from motion detection can be blew up, subsequently subtle location should be done again by integration chart. Last, the object from subtle location and the image templates from each angle should be matched. When matching the templates, the features can be extracted by up and down decomposition to the objects. The object with best match and over the threshold can be set as the object to be tracked further. Experiments show that the amount of computation can be reduced greatly and the speed can be improved by the route scheme. The tracking precision has been improved by the precision location of integration chart and the match of multi-template. The method in this paper can become a new method to effectively tracking the specific object of multi-viewpoint video.
Keywords :
feature extraction; image matching; image segmentation; motion estimation; object tracking; video retrieval; down decomposition; feature extraction; image templates; image thresholding; integration chart; motion detection; multitemplate decomposed route scheme; multitemplate decomposed tracking method; multiviewpoint video; object tracking method; precision location; subtle location; template matching; up decomposition; video retrieval; Cameras; Feature extraction; Image color analysis; Path planning; Target tracking; Vehicles; image retrieval; path planning; target tracking; video retriveal;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Systems (GCIS), 2012 Third Global Congress on
Conference_Location :
Wuhan
Print_ISBN :
978-1-4673-3072-5
Type :
conf
DOI :
10.1109/GCIS.2012.24
Filename :
6449499
Link To Document :
بازگشت