The following information was submitted:
Transactions: WSEAS TRANSACTIONS ON COMPUTERS
Transactions ID Number: 31-553
Full Name: Luo Jianguo
Position: Associate Professor
Age: ON
Sex: Male
Address: School of Electronics , Jiangxi University of Finance and Economics, Nanchang 330013, China
Country: CHINA
Tel:
Tel prefix:
Fax:
E-mail address: zytong@mailst.xjtu.edu.cn
Other E-mails:
Title of the Paper: A novel robust facial expression recognition approach
Authors as they appear in the Paper: Luo Jianguo
Email addresses of all the authors: zytong@mailst.xjtu.edu.cn
Number of paper pages: 10
Abstract: Many facial expression recognition approaches have been suggested in various applications. Robustness becomes more and more important for these applications especially in security systems. This paper proposes a novel robust facial expression recognition approach based on robust minimax probability machine (RMPM). By addressing the issue of robustness with respect to estimation errors in the means and covariance matrices via a simple modification of the input data, it can guarantee that the test set accuracy (TSA) is always greater than the estimated low bound for the RMPM. The experiments illustrate that the RMPM-based recognition approach outperforms the MPM-based approach.
Keywords: minimax probability, robust, facial expression, recognition.
EXTENSION of the file: .doc
Special (Invited) Session:
Organizer of the Session:
How Did you learn about congress: luzhaogan@126.com
IP ADDRESS: 117.32.153.139
