Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection
IEEE Transactions on Pattern Analysis and Machine Intelligence
Face Recognition Using Laplacianfaces
IEEE Transactions on Pattern Analysis and Machine Intelligence
Face Recognition by Stepwise Nonparametric Margin Maximum Criterion
ICCV '05 Proceedings of the Tenth IEEE International Conference on Computer Vision - Volume 2
Graph Embedding and Extensions: A General Framework for Dimensionality Reduction
IEEE Transactions on Pattern Analysis and Machine Intelligence
Nearest-neighbor classifier motivated marginal discriminant projections for face recognition
Frontiers of Computer Science in China
Hi-index | 0.00 |
This paper develops a nonparametric marginal Fisher analysis (NMFA) technique for dimensionality reduction of high dimensional data. According to the different distributions of the training data, two classification criterions are proposed. Based on the new classification criterions, the local mean vectors with most discriminative information are selected to construct the corresponding nonparametric scatter matrices. By discovering the local structure, NMFA seeks to find a projection that maximizes the minimum extra-class distance and minimizes the maximum intra-class distance among the samples of single class simultaneously. The proposed method is applied to face recognition and is examined using the ORL and AR face image databases. Experiments show that our proposed method consistently outperforms some state-of-the-art techniques.