![](/images/graphics-bg.png)
Improved Generalized Belief Propagation for Vision Processing
Joint Authors
Li, Ming
Liu, Sheng
Wang, Zhongjie
Tong, Hanyang
Zhang, Beiwei
Chen, Sheng-yong
Source
Mathematical Problems in Engineering
Issue
Vol. 2011, Issue 2011 (31 Dec. 2011), pp.1-12, 12 p.
Publisher
Hindawi Publishing Corporation
Publication Date
2010-12-02
Country of Publication
Egypt
No. of Pages
12
Main Subjects
Abstract EN
Generalized belief propagation (GBP) is a region-based belief propagation algorithm which can get good convergence in Markov random fields.
However, the computation time is too heavy to use in practical engineering applications.
This paper proposes a method to accelerate the efficiency of GBP.
A caching technique and chessboard passing strategy are used to speed up algorithm.
Then, the direction set method which is used to reduce the complexity of computing clique messages from quadric to cubic.
With such a strategy the processing speed can be greatly increased.
Besides, it is the first attempt to apply GBP for solving the stereomatching problem.
Experiments show that the proposed algorithm can speed up by 15+ times for typical stereo matching problem and infer a more plausible result.
American Psychological Association (APA)
Chen, Sheng-yong& Tong, Hanyang& Wang, Zhongjie& Liu, Sheng& Li, Ming& Zhang, Beiwei. 2010. Improved Generalized Belief Propagation for Vision Processing. Mathematical Problems in Engineering،Vol. 2011, no. 2011, pp.1-12.
https://search.emarefa.net/detail/BIM-470497
Modern Language Association (MLA)
Chen, Sheng-yong…[et al.]. Improved Generalized Belief Propagation for Vision Processing. Mathematical Problems in Engineering No. 2011 (2011), pp.1-12.
https://search.emarefa.net/detail/BIM-470497
American Medical Association (AMA)
Chen, Sheng-yong& Tong, Hanyang& Wang, Zhongjie& Liu, Sheng& Li, Ming& Zhang, Beiwei. Improved Generalized Belief Propagation for Vision Processing. Mathematical Problems in Engineering. 2010. Vol. 2011, no. 2011, pp.1-12.
https://search.emarefa.net/detail/BIM-470497
Data Type
Journal Articles
Language
English
Notes
Includes bibliographical references
Record ID
BIM-470497