Flavio Augusto Prieto Ortiz

Analysis and Comparison of the Cone Curvature Descriptor in Facial Gesture Recognition Tasks

Se presenta el resultado de analizar el comportamiento del descriptor de forma Cone Curvature (CC) en la tarea de reconocimiento de expresiones faciales en imagenes 3D. El descriptor CC es una representacion del modelo 3D que se calcula a partir de un conjunto de ondas de modelado para cada vertice de una malla poligonal. Se empleo la base de datos de rostros 3D (BU-3DFE), la cual contiene imagenes con 6 expresiones faciales. Con el uso del descriptor CC, las expresiones fueron reconocidas en un porcentaje promedio del 76.67% con una red neuronal, y del 78.88% con un clasificador bayesiano. Al realizar una combinacion del descriptor CC con otros descriptores como DESIRE y Spherical Spin Image, se logr´o un porcentaje promedio de reconocimiento de gestos del 90.27% y del 97.2 %, usando los mismos clasificadores mencionados previamente.

Reconocimiento de rostros y gestos faciales mediante un análisis de relevancia con imágenes 3D

The 3D face recognition aims to reduce the flaws that present the bi-dimensional based methods. This kind of recognizing method has the advantage to be invariant to illumination changes because the faces are represented as a points cloud or a 3D mesh where the most remarkable is the geometry. In this research work we present a recognizing system that uses a set of 3D shape descriptors that were selected from a relevance analysis by using the Fisher coefficients in different regions of face which are part of an anthropometric face model. A set of experiments for face, expression, and gender recognition and were performed using the relevance analysis proposed. The obtained results show that the relevance analysis offers an increasing of the performance in face recognition system.