AGOCG logo
Graphics Multimedia VR Visualisation Contents
Training Reports Workshops Briefings Index

Computer Graphics in the next 50 Years of Computing

A 2-day International Symposium took place on 29-30 October 1997 in Fraunhofer IGD, Germany, as part of the official opening of the new Fraunhofer Institute. (This official opening was reported on in Graphics & Visualization Newsletter No 57). The theme was "Computer Graphics in the next 50 Years of Computing". The objective was to look at current and future perspectives for research and development in computer graphics, multimedia, and telecommunications.

22 leaders in the field gave invited presentations under the following principal headings:

Education and Profession

Prof Steve Cunningham (California State University, USA) proposed that one of the compelling roles for computer graphics is to implement vocabularies. Images and interactions can link together computer graphics professionals and education in the following way. The vocabulary is first defined, then utilised by the profession in application domains, then further used as part of the education and training of further professionals in the field in the utilization of the new technology.

Ms Judy Brown (University of Iowa) explained how computer graphics and networks have brought about new environments for collaborative learning, research, and telemedicine. This in turn allows new styles of learning and working for the future.

Dr Christoph Hornung (Fraunhofer IGD, Germany) believed that education and training via the Internet offered a quantum improvement in opportunities for computer-based training. It offered improved qualifications for the individual and improved quality of learning for society. A key part of this is the capability for interactive group and team-based working and learning provided by the network.

Prof Scott Owen (Georgia State University, USA) predicted that by the year 2007, a PC (with typical cost under $2000) would have a processing power 100* Pentium 200, graphics 10* SGI02, 1-2 Gbytes of memory, 1 Terabyte of disk, and would support virtual environments and mobile working. The principal issue for education is to provide good quality software for remote learning, in collaboration with industry and to industry standards.

Design and Modelling

Dr Bianca Falcidenio (Istituto per la Mathematica Applicata, Italy) outlined how computer graphics and computer vision could be utilised to model, represent, and generate shapes, and incorporated into modelling systems. Limitations exist in the areas of shape analysis, model constraints, intricate features, and the design to manufacture life cycle. Higher level models and tools are needed for specification, encoding, and control.

Hierarchical methods using multiresolution representations and wavelets are being exploited by Prof Hans Peter Seidel (University of Erlangen, Germany) to compress very large amounts of data and allow efficient computation by using smoothness and coherence.

Engineering process integration for CAD, CAM, production management, production data management, and workflow management can only come about by making the underlying knowledge of engineering explicit and shareable by its participants, according to Prof Martti Mantyla (Helsinki University of Technology, Finland). Information exchange between activities and actors of engineering needs to be based on an ontological foundation.

According to Prof Ari Requicha (University of Southern California), key issues for the future are modelling beyond geometry, customization - the customer as designer, cooperative model building, interoperability, and reusability. In addition, it is important for the areas of CAD, computer graphics, and computer vision to work together on research in shape, rather than proceeding in isolation from each other.

User Interfaces

Prof David Duce (Rutherford Appleton Laboratory, UK) outlined how syndetic modelling combined a formal systems approach with human cognition to give an integrated view of human-system interaction. What users needed was a better use of more of their senses and increased user awareness by the system. Linking together the system model and the user model in the same mathematical framework enabled the conjoint behaviour of the user and the system to be represented. This gives a more accurate picture of the human/computer interaction.

"How real are virtual humans?" was a question posed by Prof Nadia Magnenat Thalmann (University of Geneva, Switzerland). Representation of emotion, facial expressions, and behaviours are key aspects for more life-like representations, whether in films or virtual worlds (e.g. for avatars representing participants). Methods for high level task description are needed in order to specify scenarios for actors and scenes. These can then be translated by intelligent software into the required lower level commands for the appropriate movement, expression, and behaviour.

"3D for All" will come via real-time PCs, down-loading of 3D models over the Internet, and standardization of data formats, according to Prof Jarek Rossignac (Georgia Institute of Technology, USA). Application areas will be 3D collaborative design, 3D multimedia documents, online training and documentation, Internet-based purchasing and contracting, and advertising.

Dr Nahum Gershon (Mitre Corporation, USA) summarised the work of the task force of the IEEE Computer Society set up to investigate "Human Centred Information Systems". Technical areas include visualization, information visualization, knowledge discovery and databases, and hypertext. The task force includes a program of work to improve user interfaces, frameworks and paradigms.

Networked Computer Graphics

Dr Detlef Kromker (Fraunhofer IGD, Germany) summarised the development of computer graphics and data networks from the 1970s to the present day. The requirements for the broadcasting industry led to the development of specialized transmission systems and infrastructures. Networks are now moving towards multipurpose capability (e.g. ISDN for speech and data, interactive TV, and Internet broadcasting), which in turn provides more capability for the end user. However, greater input/output modalities will be required in the interfaces of the future (e.g. visual, acoustic, tactile) which in turn will require continued developments in networks and services to support the integration and transmission required.

"From Personal to Transparent Computing" was the theme of Dr Michael Macedonia (Fraunhofer Center for Research in Computer Graphics, USA). Over the period 1993-96 the number of networks on the Internet had expanded from 9000 to 50,000. Fibre optic cables now linked major networks together across the world. Improved hardware and software provides support for multimedia, 3D graphics, audio, and network access. Key functions now supported include teleconferencing, high quality audio, unified messaging, broadcast video, and integrated applications (e.g. whiteboard, videoconferencing, and shared CAD). With better user-interfaces (including gesture) it will be possible to hide the computer and network latency from the user and allow seamless access to aggregations of networks and computers as an overall visual medium.

Theresa-Marie Rhyne (EPA Scientific Visualization Centre, USA) surveyed the current developments in data mining, visualization, and sharing integrated scientific information. The development of Web tools (e.g. VRML) allows collaborative sharing of visualizations between distributed users in real time. A virtual reality transfer protocol could support real-time video and interactive 3D graphics over Internet.

Dr Jim Thomas (Battelle Pacific Northwest Laboratory, USA) outlined a new human information discourse strategy for the analysis of large amounts of multimedia and multisource information. The objective is to discover, understand and communicate in ways that transcend the current limited point and click desk-top metaphor. The emphasis should be upon higher order cognitive paradigms used within the information ecosystem, and to present information through associative patterns very similar to the early visions of Vannevar Bush. Concept and insight-based systems are needed in the future. Analysts should be able to represent known information in such a way that the degree of conformity with current hypotheses and theories will be apparent.

Virtual Environments

Prof Henry Fuchs (University of North Carolina at Chapel Hill, USA) proposed an office of the future where telecollaboration used virtual reality and augmented reality to capture and display information for the users. This project is part of the NSF Science and Technology Center for Computer Graphics and Scientific Visualization. A strong sense of presence is needed for users to feel involved. Cumbersome head mounted displays are not practicable. 3D scene acquisition must be wide field of view and high resolution. The user must feel part of the distant place in real time and interacting with real people. Simulating the feeling of actually "being there" is the challenge for the future.

Figure 1and 2 Prof Bob Hopgood chairs the session on Networked Computer Graphics

Dr Roger Hubbold (University of Manchester, UK) argued for virtual environments becoming the interface of the future so that the computer was no longer directly visible in the interface to the user. Two key issues for the future are the use of perceptual measures to underpin designs (i.e. human perception of environments and what is needed to make them meet our expectations), and the semantics of virtual environments (i.e. to describe and construct virtual worlds which are lifelike and engaging to the users).

Prof Daniel Thalmann (Swiss Federal Institute of Technology, Switzerland) explained how the sense of presence was important for collaborative activities involving multiple remote users. Telepresence is the future of multimedia systems and will allow participants to share professional and personal experiences, meetings, games, and entertainment. The ultimate objective in creating realistic and believable virtual actors is to build autonomous virtual humans with adaptation, perception, and memory.

The US National Research Council has issued a report on "Modelling and Simulation: Linking Entertainment and Defence" (available on-line at under Computer Sciences in the Index of Subjects.

Prof Michael Zyda (Naval Postgraduate School, USA) outlined how these two communities are connected in these areas, and also the research directions envisaged in the report:

To cover these areas required cross-disciplinary skill sets, with expertise in modelling, simulation, virtual environments, storytelling, and content production.

Rae Earnshaw