DocumentCode :
2818773
Title :
Face recognition using maximum local Fisher Discriminant Analysis
Author :
Wang, Lei ; Ji, Hongbing ; Shi, Ya
Author_Institution :
Sch. of Electron. Eng., Xidian Univ., Xi´´an, China
fYear :
2011
fDate :
11-14 Sept. 2011
Firstpage :
1737
Lastpage :
1740
Abstract :
Compared to globality based supervised dimensionality reduction methods such as Fisher Discriminant Analysis (FDA), locality based ones including Local Fisher Discriminant Analysis (LFDA) have attracted increasing interests since they aim to preserve the intrinsic data structures and are able to handle multimodally distributed data. However, both FDA and LFDA are usually solved via a ratio trace form to approximate the trace ratio, which is the Fisher´s original objective criterion. In this paper, a novel trace optimization framework is presented to solve the original trace ratio problem. It offers an exact solution via mathematical programming and recovers Fisher´s maximal separability faithfully. The resulting maximum Local Fisher Discriminant Analysis (maxLFDA) not only inherits the merits of LFDA, but also boosts the classification accuracy in each target subspace with expected maximum trace ratio value. Experiments on a toy example and real-world face databases validate the effectiveness of the proposed method.
Keywords :
approximation theory; data handling; data structures; face recognition; image classification; mathematical programming; visual databases; classification accuracy; face databases; face recognition; intrinsic data structure preservation; mathematical programming; maxLFDA; maximum local Fisher discriminant analysis; multimodally distributed data handling; trace optimization framework; trace ratio approximation; Accuracy; Data visualization; Databases; Eigenvalues and eigenfunctions; Face recognition; Optimization; Testing; Local Fisher Discriminant Analysis; face recognition; mathematical programming; trace ratio;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Image Processing (ICIP), 2011 18th IEEE International Conference on
Conference_Location :
Brussels
ISSN :
1522-4880
Print_ISBN :
978-1-4577-1304-0
Electronic_ISBN :
1522-4880
Type :
conf
DOI :
10.1109/ICIP.2011.6115794
Filename :
6115794
Link To Document :
بازگشت