Visualization

"Vi­su­al­iza­tion is any tech­nique for cre­at­ing im­ages, di­a­grams, or an­i­ma­tions to com­mu­ni­cate a mes­sage. Vi­su­al­iza­tion through vi­sual im­agery has been an ef­fec­tive way to com­mu­ni­cate both ab­stract and con­crete ideas since the dawn of man. Ex­am­ples from his­tory in­clude cave paint­ings, Egypt­ian hi­ero­glyphs, Greek geom­e­try, and Leonardo da Vinci's rev­o­lu­tion­ary meth­ods of tech­ni­cal draw­ing for en­gi­neer­ing and sci­en­tific pur­poses." (Wikipedia) Our own re­search focus is the 3D Vi­su­al­iza­tion, mean­ing the vi­su­al­iza­tion of three di­men­sional ob­jects. Ex­am­ple ap­pli­ca­tions using 3D Vi­su­al­iza­tion are:

  • Com­puter Aided Con­struc­tion (ar­chi­tec­ture, en­gi­neer­ing, ...)
  • the vi­su­al­iza­tion of 3-di­men­sional med­ical data, for ex­am­ple from com­puter to­mog­ra­phy, 
  • the vi­su­al­iza­tion of land­scape data, show­ing ei­ther sta­tus quo or fu­ture mea­sure­ments.

Vi­su­al­iza­tion is fun­da­men­tal for many geo­met­ri­cal prob­lems. When­ever a new al­go­rithm for the cal­cu­la­tion of a 3D Voronoi Di­a­gram or me­dial axis is im­ple­mented, the very first test if the new method pro­duces the right re­sults is usu­ally op­ti­cal.
One of the first projects of the Welfenlab was the developement of a NURBS base visualization software called "Janus" ("Just Another NUrbs System"). Despite his name, the program could visualize not only NURBS surfaces but also bezier patches (and curves), triangle surfaces, point clouds or curve-on-surface-models.

3D Visualization in the Field of Medicine

A currently very hot focus of our department is the visualization of voxel data in the field of medicine. A voxel can be seen as a three dimensional analogon of a pixel. A pixel (defined by a x-, y- and a color value) describes a point on a computer monitor, while the voxel (which contains an additional z-value) describes a three dimensional point in space. Voxel data in a medical context is gained by different imaging techniques, such as computer tomography (CT) or magnetic resonance imaging (MRI).  The question remains, how such voxel data can be visualized in a human-understandable, natural way. Depending on the application it is either necessary to produce high quality images or a fast, interactive form of visualization where the user can move free around (or through) the virtual object.

Visualization of medical voxel data is often combined with the problem of segmentation, which describes the partitioning of our (three dimensional) image into multiple regions. As an exampel, a part of a voxel data set gained by a CT scan might be described as "bone", a subset of the bone might be called "jaw". Both, "bone" and "jaw" are called segments, the process of their creation (manual or automatic) is called segmentation.

Both applications, visualization and segmentation of medical data is in the focus of our research. For this purpose, we developed the sofware YaDiV ("Yet Another Dicom Viewer"), which is able to visualize, segment and register medical data (DICOM directory format). The software is currently used as a platform for our own research and is evaluated with medical scientists from the Medizinischen Hochschule Hannover (MHH). 

Visualization with Game Engines

Another focus of our visualization research is the so called "serious gaming". In the recent years, the computer gaming industry has become a large and important market and impressive amounts of money are spent on the development of new game engines. In contrast to their development costs, the price for the final product is very low compared to a professional 3D visualization/animation program.
The idea to use this potential for other purposes than gaming seems obvious.