Tele-collaboration technology, such as video-conferencing and shared desktop systems, is commonly used to enable communication and data sharing across a group of people in distant locations. This thesis argues that, as of today, linking immersive Virtual Reality (VR) interfaces over a Collaborative Virtual Environment (CVE) provides the closest resemblance of co-location, as remote people and shared objects are brought together within a shared spatial and social context.

In an increasingly global economy there is rising pressure to expand collaboration from co-located to geographically distributed groups. Currently natural human interaction is not well supported between people collaborating across a distance. This negatively impacts on the feeling and performance of collaboration. Cooperative working could be better supported by richer mediums with more natural interfaces that allow people to interact with shared objects and each other as if they were co-located. For example, intention and opinion must be communicated, while synchronously manipulating shared artefacts. Transferring the straightforwardness of such collaboration onto distributed teams is challenging. Various forms of teleconferencing systems attempt to offer such support, yet they have difficulties with sharing objects and the direct social response this involves when participants interact with those objects. This work demonstrates that a collaborative virtual environment (CVE) can assist such cooperation and that immersive displays are of greater help compared with the traditional desktop interfaces to bring us closer to replicating a face-to-face interaction. The effectiveness of application of this technology depends on a complex set of factors that determine the efficiency of collaboration. This work examines these factors and their interrelationships within the framework of a taxonomy focussed on supporting closely-coupled collaboration using immersive CVEs.

No description available.

No description available.

No description available.

In contrast to virtual reality, which immerses the user in a wholly computergenerated perceptual environment, augmented reality systems superimpose virtual entities on the user's view of the real world. This concept promises to fulfil new applications in a wide range of fields, but there are some challenging issues to be resolved. One issue relates to achieving accurate registration of virtual and real worlds. Accurate spatial registration is not only required with respect to lateral positioning, but also in depth. A limiting problem with existing optical-see-through displays, typically used for augmenting reality, is that they are incapable of displaying a full range of depth cues. Most significantly, they are unable to occlude real background and hence cannot produce interposition depth cueing. Neither are they able to modify the real-world view in the ways required to produce convincing common illumination effects such as virtual shadows across real surfaces. Also, at present, there are no wholly satisfactory ways of determining suitable common illumination models with which to determine the real-virtual light interactions necessary for producing such depth cues. This thesis establishes that interpositioning is essential for appropriate estimation of depth in augmented realities, and that the presence of shadows provides an important refining cue. It also extends the concept of a transparency alpha-channel to allow optical-see-through systems to display appropriate depth cues. The generalised theory of the approach is described mathematically and algorithms developed to automate generation of display-surface images. Three practical physical display strategies are presented; using a transmissive mask, selective lighting using digital projection, and selective reflection using digital micromirror devices. With respect to obtaining a common illumination model, all current approaches require either . prior knowledge of the light sources illuminating the real scene, or involve inserting some kind of probe into the scene with which to determine real light source position, shape, and intensity. This thesis presents an alternative approach that infers a plausible illumination from a limited view of the scene.

Distributed Virtual Environments (DVEs) provide a mechanism whereby dispersed users can interact with one-another within a shared \'irtual world. DVEs commonly allow users to interact with one-another in ways analogous to the real-world, e.g. mimicking Newtonian physics. A scalable DVE should enable large numbers of users to participate simultaneously, regardless of the In geographical location and hardware configurations of individual users. addition, these users should perceive a mutually-consistent virtual world in which each user perceives a consistent series of events in real-time. Collision detection and response is a fundamental requirement of most virtual environments and simulations. It is a computationally-expensive operation which must be perfonned at frequent intervals in all virtual environments which simulate the motion of solid objects. Collision detection has received large amounts of research interest and as a result a number of efficient collision detection algorithms have been proposed. However, these collision detection approaches are designed to detect collisions efficiently in simulations run on a single machine and are not capable of overcoming problems associated with scalability and consistency, which are of paramount importance in DVEs. This thesis presents a new collision detection approach, tenned distributed collision detection, which provides high-levels of scalability, consistency and responsiveness. This thesis presents the algorithms and theory which underpin the distributed collision detection approach and provides experimental results demonstrating its scalability and responsiveness.

Interest management is a widely used term within the area of virtual environments. It is so widely used that there even exist many synonyms for the concept. Thus both the terminology, and meaning of the concept are currently not well defined. The typical aim of interest management techniques within virtual environments has been to increase scalability. However, this thesis argues that the concept of interest management should not be so tightly coupled with the goal of scalable virtual environments, but be a concept in its own right, i.e. the management of interests. The main focus of this thesis is the representation of expressions of interest. The various techniques for expressing interest are surveyed and evaluated, providing the basis for the research into a suitable representation. This representation is achieved in two stages. The first part of this thesis introduces a novel dynamic interest management technique based upon set theory. It describes how it is expressive enough to implement most of the static interest management techniques currently available such as categorisation, locales, and interacting locales. By de-coupling the logic that implements these interests from the virtual environment, it can also describe how interests can be changed during the virtual environment's execution, thus making the technique dynamic. Enforcing and denying interests is also considered, allowing for the enforcement of interests integral to the requirements of the virtual environment. An example of this is denying the user the ability to be interested in artefacts that aren't visible. The new approach presented is implemented with SQL, and evaluated. The second part of this thesis focusses on the limitations of using SQL as an implementation language, focussing on issues of readability and succinctness and a lack of any abstraction mechanisms. Overcoming these limitations is treated as the primary design goal for a new domain specific language for representing interests. The thesis introduces this language, Wish, and evaluates it within the domain, demonstrating that it is as expressive as SQL yet is more readable, conceptually succinct and allows for arbitrary abstraction of complexity.

Context: Current visualisation systems are not designed to work with the large quantities of data produced by scientists today, they rely on the abilities of a single resource to perform all of the processing and visualisation of data which limits the problem size that they can investigate. Objectives: The objectives of this research are to address the issues encountered by scientists with current visualisation systems and the deficiencies highlighted in current visualisation systems. The research then addresses the question:” How do you design the ideal service oriented architecture for visualisation that meets the needs of scientists?” Method: A new design for a visualisation system based upon a Service Oriented Architecture is proposed to address the issues identified, the architecture is implemented using Java and web service technology. The implementation of the architecture also realised several case study scenarios as demonstrators. Evaluation: Evaluation was performed using case study scenarios of scientific problems and performance data was conducted through experimentation. The scenarios were assessed against the requirements for the architecture and the performance data against a base case simulating a single resource implementation. Conclusion: The virtualised visualisation architecture shows promise for applications where visualisation can be performed in a highly parallel manner and where the problem can be easily sub-divided into chunks for distributed processing.

No description available.