Intelligent virtual character systems

点击翻译按钮获取中文摘要

Recent developments in multimedia hardware and software have made it possible for many systems to incorporate audio-visual components. On the other hand, the human face as a communication object plays a major role in our interactions. This demands a special focus on computer-generated face animation as a specific area of research in multimedia systems and computer graphics. Among possible applications, communication, web-based services, interactive agents, computer games, and visual effects in movies can be mentioned. Our primary area of research is social user interfaces, i.e. a certain category of human-computer interaction where computers can show socially meaningful, verbal or non-verbal, behaviours such as gestures and expressions. This Discovery Grant is to support our long range research project - a comprehensive environment for interactive 2D and 3D face animation which uses a knowledge-based approach. This approach gives us a multi-dimensional parameter space language of hierarchical facial expressions, emotions and behaviours; creating a facial environment that uses a natural and complex knowledge representation of the face (and face expressions). To achieve these goals our development system uses component-based streaming architecture, and Face Modeling Language (FML), an XML-based language compatible with MPEG-4 XMT framework, and is called iFace.  iFACE is based on the idea of Face Multimedia Object (FMO), designed for a variety of face-centric applications.