I am part of the team in which Prof Peter Wood is the principal grantholder on ``Co-operative Data Visualization". Thus my research is primarily in accord with those aims. My contribution is in the fields of Human-Computer Interaction and Computer Graphics. I work at a high conceptual level (as opposed to the system based level).
This programme undertakes research at the forefront of using the latest computer-based techniques to assist people in making sense of huge volumes of complicated data. As such, we develop visualization toolkits, innovative 3D-graphics algorithms, Java-based tele-teaching and computer-supported co-operative work environments, and visual interfaces to systems such as databases and the World Wide Web.
Recent advances in computer and communications technology enable organizations to acquire, generate, store, and communicate large amounts of data. This data often contains information that can critically influence the effective operation of the organization. However, frequently it is impossible to use the information because it is difficult to find the critical information, or because the relationship between specific data items and the ``big picture'' is unclear.
We work on the premise that neither ``artificial intelligence'' nor other foreseeable enhancements to data processing will solve the problem: for the medium term at least, the making decisions in the face of complexity will be done by people. Instead of concentrating only on improving data processing, we also have to present data to people more effectively. Vision is the primary way in which people perceive the world, and only a visual interface can provide the necessary bandwidth to communicate complex scenarios to people.
We propose to build visualization tools that will enable organizations with large amounts of critical data to extract useful information from this data, quickly and effectively. The system will be distributed across multiple workstations and servers, will encapsulate access to databases, and will allow the users to collaborate in data exploration even while geographically separated. Example applications of the proposed system include the insurance industry, the financial sector, the manufacturing sector, and the information industry. We already work with industry in all these areas.
The programme has broadened to what might be called co-operative problem solving environments. In addition to co-operative data visualization, it now also includes projects on tele-teaching and tele-medicine.
The major research direction is in Co-operative Data Visualization. Specific projects are the following:
We investigate techniques to allow scientists to co-operate in research projects by means of shared collaborative visualization systems. The research can cover the development of tools and systems to support co-operation (e.g., adapting the new VRML standard for the in-house iIsh system), or, the project can explore ways in which computers can assist in consensus building and co-operative work during the visualization task.
The emphasis here lies on evaluating approaches used and developing and testing systems and interfaces for visualization of abstract multi-dimensional data. Finding appropriate metaphors and techniques that make complex results available to human understanding is one of the most challenging areas.
Object-oriented methods are well suited to graphical applications but suffer from a number of drawbacks. An important current issue is the language and standards that are suitable for distributed 3-D interaction. The current de facto standard is VRML can serve as a point of departure. This project will have to consider how constraints and other declarative features, can be integrated in an object-oriented (imperative) world. This work is a theoretical investigation but should lead to demonstrable practical benefits.
This project investigates realistic image synthesis for computer-aided design. The work will be done in close co-operation with the Interactive Systems Department of the Dutch National Centre for Research in Mathematics and Computer Science (CWI, Amsterdam). The CWI has developed a new architecture for personal workstations (with custom VLSI components) and research is now being proposed for a second generation system. A major issue will be incorporating texture mapping into a real-time graphics architecture.
Wavelet techniques are combined with considering human perceptual characteristics to develop efficient image manipulation and rendering algorithms and algorithms aimed at hardware implementation.
The research is based on the fact that Animal Trackers are a vast untapped source of information on animal behaviour and wildlife management. This project builds on the recent honours project that established the feasibility of using pen based computers coupled with Global Positioning System (G.P.S.) to assist trackers in recording their observations. The net phase is to refine the user interface for the trackers and to develop a full data visualization system. This project has attracted a great deal of publicity and at the time of writing we are confident of securing additional funding for equipment.