On dimensionality reduction for client specific discriminant analysis with application to face verification

Authors:
Xiaojun Wu;Kittler Josef;Jingyu Yang;Messer Kieron;Shitong Wang;Jieping Lu
Affiliations:
Jiangsu University of Science and Technology, Zhenjiang, P R China;CVSSP, University of Surrey, Surrey, UK;School of Information, Nanjing University of Science & Technology, Nanjing, P R China;CVSSP, University of Surrey, Surrey, UK;School of Information and Control, Southern Yangtze University, Wuxi, P R China;Jiangsu University of Science and Technology, Zhenjiang, P R China
Venue:
SINOBIOMETRICS'04 Proceedings of the 5th Chinese conference on Advances in Biometric Person Authentication
Year:
2004

Citing 8
Cited 0

Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection

IEEE Transactions on Pattern Analysis and Machine Intelligence
A new algorithm for generalized optimal discriminant vectors

Journal of Computer Science and Technology
A New Direct LDA (D-LDA) Algorithm for Feature Extraction in Face Recognition

ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 4 - Volume 04
An Optimal Set of Discriminant Vectors

IEEE Transactions on Computers
On the Discriminant Vector Method of Feature Selection

IEEE Transactions on Computers
Eigenfaces for recognition

Journal of Cognitive Neuroscience
Proceedings of the 4th international conference on Audio- and video-based biometric person authentication

AVBPA'03 Proceedings of the 4th international conference on Audio- and video-based biometric person authentication
Face recognition using LDA-based algorithms

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we propose a study on dimensionality reduction for client specific discriminant analysis with application to face verification A new algorithm of face verification based on client specific discriminant analysis is developed Two aspects of improvement are made in the new algorithm First, a dimensionality reduction based on the between-class scatter matrix is introduced which is more efficient than that based on the population scatter matrix The second improvement lies in the use of a new Fisher criterion function which is introduced in order to reduce the computational complexity of the client specific discriminant analysis problem The experimental results obtained on the internationally recognized facial database XM2VTS using the Lausanne protocol show the effectiveness of the proposed method.