Chapter 4

Visual Emotion Recognition: Status and Key Issues 1

4.1. Introduction

In this chapter we will focus on the automatic visual recognition of facial emotional expression. This kind of visual recognition requires the presence of visible signs indicating an emotion. However, the term emotion in itself can be misleading. For example, the Oxford English Dictionary defines emotion as a strong feeling, such as joy or anger, which is an instinctive feeling as distinguished from reasoning or knowledge. A facial expression, however, is a change in the face due to a voluntary or involuntary movement of several muscles demonstrating an emotional state and acting as a reflection or regulator of discourse [KAI 01]. An emotion can therefore cause (or not) a facial expression and vice versa. Since these two terms are not equivalent, we will use the term facial expression for these semantic reasons.

Interest in automatic facial recognition has grown sharply in recent years due to the multitude of potential applications for emotion recognition. Numerous conferences have held special sessions on the subject: International Conference on Image Processing (ICIP), International Conference on Automatic Face and Gestures Recognition, International Conference on Multimedia Interaction (ICMI), Human Computer Interaction (HCI) and a network of excellence [NOE 04] has also been set up to focus on this area of research.

The appearance of a facial expression is a manifestation of an emotional state ...

Get Emotion-Oriented Systems now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.