Class-specific subspace discriminant analysis for high-dimensional data - Archive ouverte HAL Accéder directement au contenu
Chapitre D'ouvrage Année : 2006

Class-specific subspace discriminant analysis for high-dimensional data

Résumé

We propose a new method for discriminant analysis, called High Dimensional Discriminant Analysis (HDDA). Our approach is based on the assumption that high dimensional data live in different subspaces with low dimensionality. We therefore propose a new parameterization of the Gaussian model to classify high-dimensional data. This parameterization takes into account the specific subspace and the intrinsic dimension of each class to limit the number of parameters to estimate. HDDA is applied to recognize object parts in real images and its performance is compared to classical methods.

Dates et versions

hal-00104065 , version 1 (05-10-2006)

Identifiants

Citer

Charles Bouveyron, Stéphane Girard, Cordelia Schmid. Class-specific subspace discriminant analysis for high-dimensional data. Craig Saunders; Marko Grobelnik; Steve Gunn; John Shawe-Taylor. Subspace, Latent Structure and Feature Selection, Statistical and Optimization, Perspectives Workshop, SLSFS 2005, Springer, pp.139-150, 2006, Lecture Notes in Computer Science volume 3940, ⟨10.1007/11752790_9⟩. ⟨hal-00104065⟩
334 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More