Statistical and neural methods for vision-based analysis of facial expressions and gender

Abstract

It is a prerequisite for the successful application of service robots in human dominated domains to have intuitive and natural man-machine interfaces. In this context, it is desirable to extract information, such as the gender, age, or emotional state (via facial expressions) about the user to aid in communication. We developed a method to estimate the facial expression and the gender of a person based on statistical data analysis and neural classifiers.