There are three subtopics in Visualization at KTH.
Visualization and Data Analysis
Analyzing large and high-dimensional data sets is a challenging task and is ideally carried out using sophisticated tools which make it possible to concentrate on the most relevant information and to automate the analysis. These goals can be achieved using feature-based methods, which foster target-oriented studies of the most important aspects of a data set.
We develop algorithms and mathematical models to analyze data sets from a variety of different domains such as computer graphics, fluid dynamics, planetology, and cell biology. A basis of our work is discrete Morse theory, as it allows robust, parameter-free, and topologically consistent computations. Our goal in this area is to develop solutions to the open problems of discrete Morse theory in order to make discrete approaches applicable for a large number of practical applications.
Interactive Virtual Characters and Computer Game Technologies
We work with the simulation and visualization of virtual individuals, groups and crowds which are significant in numerous application areas. These range from the creative industries, where virtual actors are used to represent humans in films and non-player characters (NPCs) in computer games, to more serious applications, such as the cost-effective design of safe and pleasant physical environments. Core research in this domain shares many relationships with research in multi-agent systems and social and mobile robotics and is therefore also relevant to a broad variety of application areas.
Our work concerns interaction at three scales: crowd, group and individual. These relate to the automatic generation of navigation trajectories for crowds, formation structures for groups, and full-body and facial behaviours for individuals. Behaviour generation is based on a variety of methods. These range from the detection and mapping of expressive states from humans through visual means, to more detailed internal simulation models encompassing computational synthetic perception, visual attention and memory. Human perception has a central role in our studies, helping to validate the plausibility of behaviours from the human perspective and inform further modelling efforts. We make strong use of computer game technologies and the power of graphics processing in all of our research. GPUs, game engines, middleware and assets all represent effective tools for the efficient creation of compelling real-time interactions with virtual characters and in situ visualization of simulation results.
The focus of our research is human-centered visualization systems. We currently explore a spectrum of mixed-reality, multimodal, immersive interactive analytics where, for instance, collaborating groups of users see ultra-high resolution, wall-sized displays of data at the same time as they hear and touch relevant aspects of the data. We perform task-centric analysis using user-centered design paradigms employing a spectrum of technologies, from heads-up virtual reality displays, mobile augmented reality interfaces and wall stereoscopic virtual holograms without heads-up displays.
Being task-centric development, we also focus on online data visualization platforms such as D3js and webGL development and testing. Finally, we conduct research in education of Information Visualization, including constructively aligned participation in open national student research and academic competitions.