A Realistic Video Avatar System for Networked Virtual Environments

With the advent of collaborative virtual reality applications, there is a need for representing users with a higher degree of realism for better immersion. Representing users especially with facial animation in an interactive collaborative virtual environment is a daunting task. It is observed that video directly used for rendering avatars achieves higher degree of realism in comparison to a generic model that may have been modified using extracted parameters from the video data.