Recent Progress in Brain and Cognitive Engineering by Seong-Whan Lee Heinrich H. Bülthoff & Klaus-Robert Müller

Author:Seong-Whan Lee, Heinrich H. Bülthoff & Klaus-Robert Müller
Language: eng
Format: epub
Publisher: Springer Netherlands, Dordrecht

Keywords

Facial expressionsCross-cultural psychologyEmotionsConversational expressionsGraph models

7.1 Introduction

Human communication can be divided into verbal and non-verbal signals. In the case of non-verbal signals, the human face plays a key role: the face itself conveys the person’s identity and additional kinds of attributes such as attractiveness, intelligence, and trustworthiness, for example. Importantly, when the face starts to move, facial expressions are produced that convey information about one’s feelings, emotions, or intentions – the face starts to communicate. This “language of facial expressions” allows for rich and efficient interaction between people and forms one of the most important parts of non-verbal communication. It is therefore not surprising that there is a considerable amount of research about facial expressions in various fields, focusing on both cognitive aspects (that is, the investigation of how humans perceive, process, and use facial expressions) and computational aspects (that is, the investigation of how one may teach computers to understand and react to human facial expressions).

First, concerning cognitive aspects, an interesting debate in the field has been whether certain kinds of facial expressions may be “universal” signals of communication. For example, some cross-cultural studies have found evidence for highly robust interpretation and recognition of a certain group of six emotional facial expressions (the so-called “universal”, or basic facial expressions: anger, disgust, fear, happiness, sadness, and surprise). However, other studies have shown important differences even for these six expressions across different cultural backgrounds. In addition to this debate, another important aspect that has so far been largely neglected is that human communication consists of a much broader variety of signals that do not only transport emotional, but also cognitive and other socially-regulated intentions. In order to better understand how humans perceive and process facial expressions, detailed investigations of the perceptual and cognitive aspects of facial expression processing, taking into account the broad repertoire of expression signals as well as cross-cultural contexts, are necessary.

Secondly, concerning computational aspects of facial expression research, many frameworks for automatic and efficient human-computer-interaction have been proposed by computer vision researchers. So far, however, the available frameworks are not capable of recognizing more than the basic (six) emotional facial expressions. If we want to interpret the much larger range of general conversational expressions, we need to address some crucial obstacles: first, there is a large variability across individuals in expressing certain intentions, which represents a challenge for efficient modeling of expression categories. Second, conversational signals are often conveyed using highly subtle facial movements, which is another challenge for automatic facial feature tracking algorithms. Therefore, we need to develop novel computational frameworks that are capable of dealing with these issues such that we may interpret and process the full range of human communication.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.