Issues on Displaying 3D Data for Scientific Visualization
Institute of Computer Graphics
Vienna University of Technology
In the field of computer graphics many techniques and algorithms have been developed to render three dimensional data on two dimensional displays, often trying to do this as accurate as possible. In scientific visualization, however, generating images which provide maximal insight into three-dimensional data is more important than rendering an image in a photo-realistic way.
Particularly when depicting large surfaces occlusion becomes a major problem. Objects that lie behind or within these surfaces cannot be seen. Using transparency improves the situation but still there are problems: transparent surfaces provide little cues on their shape, especially the shape of the front-facing portion usually cannot be perceived accurately. Further, multiple superimposed transparent layers cannot be distinguished at all and it is almost impossible to estimate the distance in depth between two such layers. Small, sparse, opaque marks on a transparent surface facilitate the perception of both the shape and depth of a layered, transparent surface without impairing too much the visibility of objects located behind it. By taking into account the curvature of the surface when applying these marks, the perception of the shape of the surface itself can even be enhanced.
Another important issue is what part of the data should be depicted not to obtain pictures which are overloaded and confusing. Especially when visualizing three-dimensional flow using stream surfaces it is not only necessary to consider how to render these surfaces but also which of the infinite possible surfaces to choose . Also the distribution of streamlines is not trivial .
When depicting lines in three dimensional space in general, occlusion is also a problem , but more problematic is that these lines when not in motion tend to be flattened, they appear to be in one plane . The use of common lighting models and the thereby achieved shading effects give important cues for the spatial perception, but there are also other approaches that try to enhance the perception of lines in three dimensional space.