Graphics: Visual and Interactive Computing

• VIC  | 
• Volume Graphics  | 
• Data Visualization  | 
• Interactive  | 
• Intelligent  | 
• Members

Data Visualization

Academics: M. Chen, R. S. Laramee, M. W. Jones, B. Mora

Collaborators: Stuttgart (Germany), Bangor, Leeds, Manchester (UK), Simon Fraser (Canada), Swansea Psychology (UK), Swansea Scientific Computing Group (UK)

The most significant contribution in the field of visualization made by Swansea is the introduction of video visualization as a visual analytics technique for extracting meaningful information from video sequence. Recently, in collaboration with Stuttgart (Germany), Simon Fraser (Canada) and Swansea Psychology, the group formulated the notion of visual signatures in video visualization, developed combined volume and flow visualization techniques for displaying visual signatures, and provided a set of convincing results to demonstrate that human observers can recognize motions from static visual representations of videos. This opens a new frontier in the field of visualization.

Dr. Laramee, who joined the group recently, is among the most noticeable young scientists in the field of flow visualization, and has made a number of important advances in developing flow visualization techniques, including direct and interactive geometric techniques for 2D and 3D vector fields, dense and texture-based techniques for depicting flows on surfaces, and techniques for extracting and visualizing swirl and tumble motion. Much of the work was directly applied to industrial applications.

The combination of rapid development of commodity graphics hardware techniques and explosive increase of data to be visualized raises a fundamental question about the future computing infrastructure for visualization, which is the research focus of a major collaborative project involving Swansea, Bangor, Leeds, Manchester, and Utah. The group proposed to address the complexity in managing data, resources and interaction in a visual supercomputing infrastructure by adopting autonomic computing evolutionarily. It developed a simulation environment for modelling and analysing different infrastructures including those based on future technologies and those which are mission-critical and unsuitable for experimentation. It also developed a prototype infrastructure supported by an agent-based visualization framework.

The group also worked in collaboration with visualization users in medicine, bioinformatics, physics, forensic science, and computational fluid dynamics. It has developed a number of application-driven techniques, such as fibre-bundle tracing in heart modelling, and comparative visualization for computational fluid dynamics.