DLDA/QR: a robust direct LDA algorithm for face recognition and its theoretical foundation

Authors:
Yu-Jie Zheng;Zhi-Bo Guo;Jian Yang;Xiao-Jun Wu;Jing-Yu Yang
Affiliations:
Department of Computer Science, Nanjing University of Science and Technology, Nanjing, P.R. China;Department of Computer Science, Nanjing University of Science and Technology, Nanjing, P.R. China;Department of Computing, Hong Kong Polytechnic University, Kowloon, Hong Kong;School of Electronics and Information, Jiangsu University of Science and Technology, Zhenjiang, P.R. China;Department of Computer Science, Nanjing University of Science and Technology, Nanjing, P.R. China
Venue:
PAKDD'07 Proceedings of the 11th Pacific-Asia conference on Advances in knowledge discovery and data mining
Year:
2007

Citing 11
Cited 0

Introduction to statistical pattern recognition (2nd ed.)

Introduction to statistical pattern recognition (2nd ed.)
Optimal discriminant plane for a small number of samples and design method of classifier on the plane

Pattern Recognition
Automatic recognition and analysis of human faces and facial expressions: a survey

Pattern Recognition
Optimal Fisher discriminant analysis using the rank decomposition

Pattern Recognition
Using Discriminant Eigenfeatures for Image Retrieval

IEEE Transactions on Pattern Analysis and Machine Intelligence
Matrix computations (3rd ed.)

Matrix computations (3rd ed.)
Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection

IEEE Transactions on Pattern Analysis and Machine Intelligence
Face recognition: A literature survey

ACM Computing Surveys (CSUR)
KPCA Plus LDA: A Complete Kernel Fisher Discriminant Framework for Feature Extraction and Recognition

IEEE Transactions on Pattern Analysis and Machine Intelligence
A Two-Stage Linear Discriminant Analysis via QR-Decomposition

IEEE Transactions on Pattern Analysis and Machine Intelligence
Effective classification image space which can solve small sample size problem

ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 02

Quantified Score

Hi-index	0.00

Visualization

Abstract

Feature extraction is one of the hot topics in face recognition. However, many face extraction methods will suffer from the "small sample size" problem, such as Linear Discriminant Analysis (LDA). Direct Linear Discriminant Analysis (DLDA) is an effective method to address this problem. But conventional DLDA algorithm is often computationally expensive and not scalable. In this paper, DLDA is analyzed from a new viewpoint via QR decomposition and an efficient and robust method named DLDA/QR algorithm is proposed. The proposed algorithm achieves high efficiency by introducing the QR decomposition on a small-size matrix, while keeping competitive classification accuracy. Experimental results on ORL face database demonstrate the effectiveness of the proposed method.