A model for exploring virtual reality environments

In current dynamic data models for VR environments, the focus has been the graphical presentation of objects and an efficient running of animations. In such models the objects' semantics are almost always assumed by the observer and derived from the context of the application. in this paper we introduce a model to represent the evolution of VR Objects' semantics over time. Based on this model, a new classification of virtual reality objects is proposed that captures semantics relative to the existence, visibility, and activity of objects. A set of six operations is presented to model the possible states of VR Objects. This paper also presents an animation model with high-level abstractions of objects' behaviors and a rich set of operations to manipulate such behaviors. With such tools, users are able to create their own views of the environment, gain insight, and discover relationships among dynamic phenomena. This paper, therefore presents a rationale for modifying object semantics under manipulations by an observer of the object behavior. (C) 2003 Elsevier Ltd. All rights reserved.