Exploratory Sparse Models for Face Classification
Nicholas P. Costen, Martin Brown, John C. Dalton · 2003
In this paper, a class of sparse regularization methods are considered for developing and exploring sparse classifiers for face recognition. The sparse classification method aims to both select the most important features and maximize the classification margin, in a manner similar to support vector machines. An efficient process for directly calculating the complete set of optimal, sparse classifiers is developed. This set can be explored in order to understand the sensitivity of feature selection process to small parametric changes. We show that this method can be used to construct a useful classification hyper-plane for faces represented via an appearance model. In addition, the stability of the classification process is explored and the incorporation of prior knowledge about the importance of individual features is considered. 1