The intensity of perceived emotions in 3D virtual humans

AS Shaarani, Daniela Romano
in Proceedings of the International Joint Conference on Autonomous Agents and Multiagent Systems, AAMAS, Conference paper (text)

Abstract

Synthetically generated 3D humans often fail to express a full range of emotions or present different levels of the same type of emotion. Transcending the facial expression, what should a happy synthetically generated human look like? What about a slightly happy or ecstatically happy? This paper reports a study aimed at identifying the appropriate bodily expressions for various emotions in 3D human-like figures at varying emotional strength. Thirty-six volunteers were asked to discriminate and categorize thirty cards with static poses of 3D human-like characters into the Ekman's six basic categories of emotions. This is to judge the compatibility of each posture in relation to each category and to rate their level of emotion within the group. Copyright © 2008, International Foundation for Autonomous Agents and Multiagent Systems (www.ifaamas.org). All rights reserved.