The modelling of affective behaviour and appropriate bodily expression to make synthetic characters more believable becomes important in many types of applications such as games, story telling, training environment, virtual therapy and interactive conversational agents. This paper describes report on the results of an experiment designed to study the empathy inspired to users when looking at static human images. An experiment was carried out with 36 participants who were asked to categorized 30 cards of static human images into six basic categories of emotions and using the rating scale, subjects judge the compatible of each posture into each emotional category. Emotional categories are differed in term of groups and level of emotional for every posture. The approach of the study is to find out which human posture of body movement represent specific emotion and the state of emotional level represent by each posture.
Ahmad S. Shaarani, Daniela M. Romano