Abstract
This paper addresses the problem of soft controlling an avatar's facial expressions by its user's facial expressions. For example, when a user laughs, the avatar also gives a laughing expression but the avatar does not necessarily mimic the user's expression exactly. The major challenge here is how to recognize the user's facial expressions in real time. Unlike conventional approaches trying to recognize universal expressions independent to individuals, we use a personalized scheme to recognize facial expressions. We represent facial movements by facial motion graph (FMG), which is based on feature points and their relations. Expression discrimination is achieved by analyzing the similarity between the new FMG and the person's FMG models of known expressions using continuous dynamic programming. Experiments show excellent accuracy and near real time speed of the system.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
