Your face is one in a million (actually, more like six billion), but that's the problem! Face recognition is an important human skill, allowing us to quickly identify people by sight so we can easily recognize family and friends and identify people we don't know as strangers.
While each of our faces is unique, many also share common traits. So while humans can recognize each other easily, it is very difficult for a robot. However, if robots learn to recognize individual faces, they could make deliveries to a specific group of people, perform security checks, or find their owner in a crowd. Gesture recognition, the ability to recognize facial expressions that indicate emotional states, such as curiosity, happiness, confusion, or anger, is an important step in developing robotic intelligence. IBM's Attentive Environments, a robot featured in roboworld™, can track your face, recognize changes in your expression, and even mimic the faces you make!
Developed by IBM, this robot demonstrates how robots can be used to gauge and evaluate human interest in products or product messaging in a store environment. Using a vision system that can track visitor movement and software capable of interpreting visual focus and recognition of facial gestures, Pong combines the technology of robotics with real-world marketing and sales efforts as an analysis tool.
The Human Computer Interaction Institute is an interdisciplinary community of students and faculty at Carnegie Mellon University. This community is dedicated to research and education in topics related to computer technology in support of human activity and society.