Supervised Hybrid Expression Control Framework for a Lifelike Affective Avatar
MetadataShow full item record
The use of an avatar-enabled application has been rapidly growing over the last decade as it promises more natural computer interaction with advanced technologies in various domains. Furthermore, recent research efforts towards natural and affective avatar capabilities have become more prevalent in the field. However, developing such application still remains very hard and time-consuming task. This is mainly because a believable avatar model intuitively aims to mimic a real human including realistic appearance and wide spectrum of complex behavior. This thesis presents a high quality visualization method and a behavior-modeling framework that can enhance user experience with an autonomous avatar and eventually achieve the goal of an avatar as a means of natural lifelike computer interface. A hybrid behavior modeling technique sets the middle ground to orchestrate both rule-based and data-driven models. Highly realistic avatar visuals with emotionally expressive behavior model offers better congruency and naturalness at the same time to increase avatar believability. A user study is conducted to evaluate the perceived naturalness of the presented behavior model within an autonomous expressive storytelling context. As we establish a better tractable model for an avatar as more natural alternative computer interface, it will broaden the possibilities of our computation needs where we suffer from our limited resources.
Embedded Conversational Agent
Avatar Behavior Modeling