Letting Go: Automating The Display Of Virtual Environments

It is commonplace to present virtual world interactions as a three-dimensional display – such as may be generated using a virtual camera. Even though these ‘stealth’ or ‘magic carpet’ displays are popular, effective presentation of information in this format is a challenging problem. Unskilled operators often select viewpoints that present confusing information to viewers. This paper examines the task of the display controller with a view to developing a real-time automated 3D view of a multi-player virtual environment. Cinematic conventions are reviewed in the context of characterising the 3D communication medium. A prototype system is then presented that demonstrates how some of these issues may be addressed.