EVL PhD candidate Sangyoon Lee presents his preliminary examination for doctoral dissertation: Friday, September 2, 2011, 10AM, EVL Cyber-Commons (2068 ERF)
A Supervised Hybrid Expression Control Framework for a Lifelike Affective Avatar (SHECF) is proposed, which promotes easy design and development for an affective avatar-enabled application. 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. Even though we have to approach this problem as a whole piece, most of studies focused on only small part of a model due to its complexity. Proposed SHECF combines highly realistic avatar visuals with emotionally expressive behavior model to overcome current limitations.Recently, the state of art researches in computer graphics present realistic renderings of human face with programmable commodity graphics processing unit. However, these results are mostly focused on static model or non-interactive character animations whereas avatar researches tend to use very limited visualization capabilities. Merging two research efforts will provide better chance to surpass Moriís Uncanny Valley and to achieve more natural experience with an avatar. Another discrepancy found in literature is two distinct methodologies towards modeling human behavior such as rule-based and data-driven model. The model of rule-based system offers highly coherent behavior based on psychological theories but it lacks in subconscious or unconscious behavior. The latter method relies on large amount of rich data to extract uncertain nature of human to mimic it on avatar model but it suffers from limited depth of knowledge that is required to understand progressive causalities in our face-to-face communications. SHECF sets the middle ground to orchestrate both models and offers better congruency and naturalness at the same time to increase avatar believability.