—Support Vector Machines are used to combine the outputs of multiple entity extractors, thus creating a composite entity extraction system. The composite system has a significant...
Deborah Duong, James Venuto, Ben Goertzel, Ryan Ri...
Dimensionality reduction plays a fundamental role in data processing, for which principal component analysis (PCA) is widely used. In this paper, we develop the Laplacian PCA (LPC...
One of the major difficulties in face recognition systems is the in-depth pose variation problem. Most face recognition approaches assume that the pose of the face is known. In th...
Abstract-- This paper introduces Evolutionary Pursuit (EP) as a novel and adaptive representation method for image encoding and classification. In analogy to projection pursuit met...
This perspective paper explores principles of unsupervised learning and how they relate to face recognition. Dependency coding and information maximization appear to be central pr...