NumberII-2-p23
Title Arabic Accented Facial Expressions for a 3D Agent
Authors Amna AlZeyara (Carnegie Mellon University)
Micheline Ziadee (Carnegie Mellon University)
Majd F. Sakr (Carnegie Mellon University)
Abstract In this study we develop facial expressions for a 3D agent that has Arabic facial features and evaluate how Arabs and non-Arabs recognize Arabic facial expressions implemented on a 3D agent. Various previous studies suggest that people expect a 3D agentfs behavior to align with its features. Therefore, a 3D agent with Arabic features must display Arabic facial expressions. In this work, we identify Arabic facial expressions by taking videos of Arabs while narrating a story that invokes an emotion. We implement these expressions on the 3D agent. We then run a survey to evaluate Arabs and non-Arabsf recognition of the facial expressions. Our results show that the recognition of expressions differs slightly between Arabs and non-Arabs. The exception is disappointment which is better recognized by Arabs. The results reveal that positive expressions, when implemented on a 3D agent, are easily recognized. Further, we find that disappointment is a culturally dependent expression for Arabs.
PDF [Link]