Generalizing Dissimilarity Representations Using Feature Lines

Abstract

A crucial issue in dissimilarity-based classification is the choice of the representation set. In the small sample case, classifiers capable of a good generalization and the injection or addition of extra information allow to overcome the representational limitations. In this paper, we present a new approach for enriching dissimilarity representations. It is based on the concept of feature lines and consists in deriving a generalized version of the original dissimilarity representation by using feature lines as prototypes. We use a linear normal density-based classifier and the nearest neighbor rule, as well as two different methods for selecting prototypes: random choice and a length-based selection of the feature lines. An important observation is that just a few long feature lines are needed to obtain a significant improvement in performance over the other representation sets and classifiers. In general, the experiments show that this alternative representation is especially profitable for some correlated datasets.

Trends in Nearest Feature Classification for Face Recognition—Achievements and Perspectives

“…Our generalization consists in creating matrices D L (T,R L ) and D F (T,R F ) by using the information available at the original representation D(T,R), where subindexes L and F stand for feature lines and feature planes respectively. This generalization procedure was proposed in and (Orozco-Alzate et al, 2007a). In this section, we review our method as it was reported in the above-mentioned references but also including some results and remarks resulted from our most recent discussions and experiments.…”

A Dynamic Programming Technique for Optimizing Dissimilarity-Based Classifiers

“…Theoretically, the stronger the linear relation between two vectors, the greater the similarity of their structures. On the basis of this idea, we use the right essence of the correlation coefficient as a measurement of the structural similarity between two face vectors 8 . To obtain an optimal warping path for two objects, we first compute the correlation coefficient between two subregions (a subsequence of vectors) selected from the object matrices by scanning through them in a vertical-only way or a vertical-horizontal way.…”

“…The three major questions we encountered when designing the DBCs concern the selection of prototype subsets [3], [4], [7], [8]; the measurement of dissimilarities between object samples [9], [10], [11]; and the design of a classifier in the dissimilarity space [2], [12]. Various methods have been proposed in the literature [3], [4], [7] as a means of selecting a representative set of data that is both compact and capable of representing the entire data set.…”