Introduction

Goals Of The Automated Face Analysis

Machines may not need all of the communication skills that people need; however, there is evidence that machines will require at least some of these skills to appear intelligent when interacting with people. A relevant theory is that of Reeves and Nass (1996) at Stanford. Human-computer interaction is inherently natural and social, following the basics of human-human interaction. For example, if a piece of technology talks to you but never listens to you, then it is likely to annoy you, analogous to the situation where a human talks to you but never listens to you. Nass and Reeves conducted dozens of experiments of classical human-human interaction, taking out one of the humans and putting in a computer, and finding that the basic human-human results still holds.

In general, the aforementioned human communication skills include many modalities such as visual, and autotory channels. Among the many skills, this book concentrates on the skills that are based on visual sensing of the human eyes. Thus, the ultimate goal of the automated face analysis that researchers dream of is to catch all the visual messages that a human face brings, and to understand the meaning of them, and to automatically react to the human in an appropriate and natural way so that the human can easily communicate with any computer, robot, or machine as the human communicates with another human.