Graph embedding discriminant analysis on Grassmannian manifolds for improved image set matching
- 1 June 2011
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10636919,p. 2705-2712
Abstract
A convenient way of dealing with image sets is to represent them as points on Grassmannian manifolds. While several recent studies explored the applicability of discriminant analysis on such manifolds, the conventional formalism of discriminant analysis suffers from not considering the local structure of the data. We propose a discriminant analysis approach on Grassmannian manifolds, based on a graph-embedding framework. We show that by introducing within-class and between-class similarity graphs to characterise intra-class compactness and inter-class separability, the geometrical structure of data can be exploited. Experiments on several image datasets (PIE, BANCA, MoBo, ETH-80) show that the proposed algorithm obtains considerable improvements in discrimination accuracy, in comparison to three recent methods: Grassmann Discriminant Analysis (GDA), Kernel GDA, and the kernel version of Affine Hull Image Set Distance. We further propose a Grassmannian kernel, based on canonical correlation between subspaces, which can increase discrimination accuracy when used in combination with previous Grassmannian kernels.Keywords
This publication has 22 references indexed in Scilit:
- Kernel Grassmannian distances and discriminant analysis for face recognition from image setsPattern Recognition Letters, 2009
- Multi-Region Probabilistic Histograms for Robust and Scalable Identity Inference2009
- Optimal Local Basis: A Reinforcement Learning Approach for Face RecognitionInternational Journal of Computer Vision, 2008
- Integrating Global and Local Structures: A Least Squares Framework for Dimensionality Reduction2007
- Face Recognition with Image Sets Using Manifold Density Divergence2005
- The BANCA Database and Evaluation Protocol2003
- On combining classifiersIEEE Transactions on Pattern Analysis and Machine Intelligence, 1998
- The Geometry of Algorithms with Orthogonality ConstraintsSIAM Journal on Matrix Analysis and Applications, 1998
- Face recognition: the problem of compensating for changes in illumination directionIEEE Transactions on Pattern Analysis and Machine Intelligence, 1997
- Eigenfaces vs. Fisherfaces: recognition using class specific linear projectionIEEE Transactions on Pattern Analysis and Machine Intelligence, 1997