DocumentCode
2464927
Title
Semantic video object tracking using region-based classification
Author
Gu, Chuang ; Lee, Ming-Chieh
Author_Institution
Microsoft Corp., Redmond, WA, USA
fYear
1998
fDate
4-7 Oct 1998
Firstpage
643
Abstract
This paper introduces a novel semantic video object tracking system using backward region-based classification. It consists of five elementary steps: region pre-processing, region extraction, region-based motion estimation, region classification and region post-processing. Promising experimental results have shown solid performance of this generic tracking system with pixel-wise accuracy
Keywords
feature extraction; image classification; motion estimation; tracking; video signal processing; backward region-based classification; generic tracking system; performance; pixel-wise accuracy; region classification; region extraction; region post-processing; region pre-processing; region-based motion estimation; semantic video object tracking; Application software; Assembly; Digital signal processing; Humans; Motion estimation; Motion pictures; Software libraries; Solids; Tracking; Video signal processing;
fLanguage
English
Publisher
ieee
Conference_Titel
Image Processing, 1998. ICIP 98. Proceedings. 1998 International Conference on
Conference_Location
Chicago, IL
Print_ISBN
0-8186-8821-1
Type
conf
DOI
10.1109/ICIP.1998.999050
Filename
999050
Link To Document