Avatar Markup Language

Synchronization of speech, facial expressions and body gestures is one of the most critical problems in realistic avatar animation in virtual environments. In this paper, we address this problem by proposing a new high-level animation language to describe avatar animation. The Avatar Markup Language (AML), based on XML, encapsulates the Text to Speech, Facial Animation and Body Animation in a unified manner with appropriate synchronization. We use low-level animation parameters, defined by the MPEG-4 standard, to demonstrate the use of the AML. However, the AML itself is independent of any low-level parameters as such. AML can be effectively used by intelligent software agents to control their 3D graphical representations in the virtual environments. With the help of the associated tools, AML also facilitates to create and share 3D avatar animations quickly and easily. We also discuss how the language has been developed and used within the SoNG project framework. The tools developed to use AML in a real-time animation system incorporating intelligent agents and 3D avatars are also discussed subsequently. 1. Introduction Avatar animation is no longer a mere research topic. Many efforts in this area, combined with standardization (e.g. MPEG-4 facial and body animation), have lead to a wide use of virtual faces and bodies for the interactive web based and entertainment applications. Recently, very believable virtual humans have been seen in the movies. However, such production demands many days or even months of manual design work from highly skilled animators. On the contrary, real-time avatar animation on the web cannot be completely pre-defined with the accuracy and the artistic precision evident in the computer generated movies of today. The advent of web-based e-commerce applications demand powerful and easy to use high-level interfaces, capable of animating web-based characters, instantly and smoothly. Such an interface finds immediate applications for web-based virtual sales assistants, controlled by an intelligent agents, and face-to-face avatar interaction. In recent years, there have been several efforts for the development of such a high-level control mechanism for character animation, from IMPROV [1] to BEAT [2]. The IMPROV system suggested use of high-level scripts and tools for creation of real-time behavior based animation with non-repetitive motions and smoothness. The system allowed animators to create rules governing how the synthetic actors communicate and make decisions. More recently, BEAT (Behavior Expression Animation toolkit) used text input to create behavioral animation, using the behavioral rules defined by the animator. The toolkit …