1. Introduction

Virtual Reality (VR) is a revolutionary new technology that will allow users to interact with computers in entirely new ways. One of the great advantages of using VR technology is described by Leler (1995) as "...the high degree of synergy possible between the user (the occupant of or participant in the virtual environment) and the computer". When utilising a virtual environment the user has the ability to directly interact with and manipulate data in real time. Currently, there is a great deal of VR research being carried out indicating the diversity of the potential applications that this technology may have.

For the purposes of this report, applications related to VR as a tool for the facilitation of higher education will be considered. A detailed review of this area can be seen in Kalawsky (1995). In summary, Kalawsky indicates a number of areas where VR techniques may prove beneficial when compared with established methods. These include:

• Simulation of complex systems.
• Macroscopic and microscopic visualisation.
• Fast and slow time simulation.
• Allow high levels of interactivity.
• Allow a sense of immersion.
• Inherent flexibility/adaptability.

There are a number of possible uses of such environments in higher education i.e. as simulations, representations and training tools.

The virtual environment that is experienced by the user must be created using the necessary software. A number of different types of software may be used in developing the environment and this procedure can be broken down into three main areas:

i) Modelling

As in the real world, a virtual environment is composed of a number of different entities or models. Van Dam and Sklar (1990) define a model as "..a representation of some (not necessarily all) features of a concrete or abstract entity". Each may have different geometry (size and shape), surface appearance (texture and material type) and may behave in an individual manner (in terms of physics, dynamics and properties). In the virtual environment, however, the model can be based on a real physical entity or a totally abstracted concept. The initial stage of authoring involves creating the models that will be used in the environment. This is generally achieved using modelling tools, but CAD tools may also be used. As Kalawsky (1995) indicates "...the creation of virtual objects represents a significant investment in time", suggesting that any technique to reduce this time such as the ability to import/export models between systems or reuse them in different environments would be highly desirable.

ii) Creating the virtual environment

An environment may contain a number of models with various behaviours. The relationships between the objects may consequently become very complex. For example, if two moving objects meet in the environment what is the result? Will a collision occur when the boundaries of each model intersect? Will the two objects bounce off each other or will they coalesce? Will the shape of the objects be changed by the impact? Will their trajectories be modified following the impact? If one considers that this scenario may occur many times with many objects it is clear that creating the environment is extremely time-consuming and involved.

iii) Run-time environment

At run time the virtual environment software is executed by the host computer. This execution may occur on a number of levels. If the environment is simple, it may just allow the visualization of a model that has been created. On a more complex level it may combine the visualisation with some animation that has been attached or may even combine the geometry with behaviours and allow interaction between objects and also between the user and the environment.

1.1 Purpose of report

The purpose of this report is to provide information about the use of 3D model databases for members of the AGOCG user commumity and the wider community interested in utilising VR techniques in higher education. For the purpose of this report it is necessary to assume some basic understanding of VR techniques and it will concentrate on areas that are felt to be particularly relevant to the educational domain. It is not the intention of this report to concentrate on hardware issues and for a detailed review of this area please see Kalawsky (1993, 1995) and Youngblut, Johnson, Nash, Wienclaw and Will (1996).

The main objectives of this report are:

• To provide an overview of object modelling considerations for VR applications.
• To detail current formats and model composition.
• To detail the requirements that the VR community may have with respect to the definition and distribution of virtual models.
• To highlight the issues which must be considered when importing and exporting models.
• To identify a number of model databases that may be available on the internet and to relate these to specific modelling packages.
• To consider future developments in the area of model definition and distribution.

1.2 Scope of report

To achieve the purpose of the report described above, it will be necessary to examine several areas.

The report will initially concentrate on object modelling. Furthermore, a list of current file formats will be presented and the composition and structure will be discussed.

The requirements of the VR community will then be investigated. This will include a review of current object modelling packages, CAD systems and behaviour models. In addition, relevant data converters, translators and filters will be reviewed.

Issues arising from these requirements and the current formats that are available will then be reviewed. These issues include conversion between standards, resolution, performance issues, cost of support and intellectual property rights with respect to model ownership and liability issues.

In section 5, sources of model databases will be discussed with particular reference to the world wide web. A number of sites will be reviewed that represent a range of different modelling systems.

In the final section, future trends will be discussed. These will include attempts to standardise systems including STEP and VRML and also future developments in development tools and techniques such as 3D scanning and input of models. The need for an AGOCG model database will also be considered.

Graphics     Multimedia      Virtual Environments      Visualisation      Contents