On this page I describe opportunities for students from DAIMI and IMV to work with me on projects. Generally each project could be executed as either a masters thesis or as a study group. I you feel that any of these projects could have your interest or you have another proposal for a master’s thesis or study group, feel free to contact me. In general I can offer supervision in projects dealing with realtime computer graphics, animation (e.g. physical based), medical visualisation and simulation, as well as computer games.
Recently a rather large number of indie game productions have experimented with physics-based game play (i.e. Gish, Toribash, Crayon) and this tendency is also somewhat evident in commercial games such as Super Mario Galaxies and the upcoming Little Big Planet. In this project you would categories the existing development and research within physics-based game play and compare this to techniques of physics simulation. Spotting a potential pattern or opportunity for a new kind of physics based game play you would implement and test your idea(s).
This is a project in Cooperation with Bang & Olufsen A/S. Computer graphics research has produced many algorithms for simulating reality. In graphical user interfaces realism is not necessarily a goal, but many of these algorithms may still be very useful in the context of GUI’s. The purpose of this project is to study and categorize the literature, analyze and implement a number of key algorithms and evaluate them in the context of user interface use. See this pdf for more detail: BangOlufsenComputerGraphics.pdf.
This is a project in Cooperation with Bang & Olufsen A/S. A visual effect of smoke, possibly based on fluid dynamics or particles, could add a magic touch to a product user interface. The goal of this project is to design and implement a method for fast, high quality rendering of smoke, mist and fluorescent gasses. See this pdf for more detail: BangOlufsenComputerGraphics.pdf.
This is a project in Cooperation with Bang & Olufsen A/S. Imagine a large content model where each piece of content is represented by an image. For ease of navigation through content, fast browsing in various orders through these images is desirable. The goal of this project is to determine a good strategy for storing content data such that it can be retrieved fast and to find an optimal way of transferring image data from storage on HDD to GPU ready for rendering. See this pdf for more detail: BangOlufsenComputerGraphics.pdf.
This is a project in Cooperation with Bang & Olufsen A/S. There are many approaches to interactive user interfaces construction. From drag-drop, static 2d centric solutions such as the Windows Forms designer in Visual Studio to very flexible, full blown 3d animation packages such as 3d Studio Max or Maya. Another approach is taken by the object oriented Quartz Designer in MacOS X. The aim of this project is to do a survey of the field of user interface design tools in the context of Bang & Olufsen’s needs, and from this derive a set of requirements and possibly prototype a simple editor tool. See this pdf for more detail: BangOlufsenComputerGraphics.pdf.
A microbiologist from Aarhus University has retrieved an infrared scanning of a couple of mice fetuses - some normal, some with abnormalities. The scanning consists of a number of 2D slices, representing a 3D volume. This particular imaging modality is very noisy, and is skewed because of the pendulum motion of the scanning device. There is furthermore a "gradient" (or shadowy) quality to the image since the infrared laser is rather quickly absorbed through the tissue. In this project we would correct for the various artifacts through image analysis and work towards a 3D volume rendering or 3D reconstruction. The microbiologist would be willing to participate in the study with domain knowledge and evaluation.
The GPU is an alternative efficient and cheap computational resource available as a shared memory parallel architecture. In cooperation with the Bioinformatics Research Center we look into the algorithmic engineering within bioinformatics using the modern GPU through CUDA or CTM.
Within the general theme of serious games, we study usage of game-elements in communication, information, learning, discussion etc. Serious games can very well be provocative and challenging towards a subject, while not even be entertaining in the usual meaning of the word. E.g. simulation-games have the unique effect of challenging people to find rules of behavior, thereby voluntarily exploring complex systems that might have a counterpart in the complex systems of our everyday world. At Institute of Information Studies and Multimedia we have a unique opportunity to research these elements from many different perspectives.
In the commercial computer game industry profit is always the end goal - if not, game companies would not survive. At the University we are allowed to experiment; to question even the foundation of what makes "a game" a game. In this study we could work with e.g. user-generated contents for games, algorithmically generated elements in games, the digital aesthetics of computer games explored through prototypes, or using the visualization and interaction equipment in the Center for Advanced Visualization and Interaction..
At the Dept. of Orthodontics engineers have retrived micro-CT images of teeth-implants, specifically concerning the screw into the jaw-bone. These images are so detailed that even the hole-filled spongy bone tissue can be seen. An screw into the jaw-bone is "strongly" fixed if a large amount of contact exists between bone and implant. In this study we seek to compute an estimate of that amount through image analysis of the micro-CT images.