Future Challenges and Unsolved Problems in Multi-field Visualization - Scientific Visualization: Uncertainty, Multifield, Biomedical, and Scalable Visualization - page 198

question remains, how and whether multi-field visualization can incorporate such
dynamic changes in an intuitive and expressive way.
Ronald Peikert on Feature-based Visualization:
The challenges of multifield visualization also extend to the area of feature-based
visualization. Many useful techniques have been developed for finding inherent fea-
tures in scientific data. They typically operate on one or at most two scalar, vector
or tensor fields. In most cases, such feature detectors are not based on concepts that
easily generalize to larger multi-fields containing additional variables. A feature can
in the simplest case be represented by scalar field indicating the presence or absence
of the feature or, alternatively, a probability for the feature to be present at a given
location. But even with this simple notion of a feature, it is not clear how to combine
a large number of them in a single visualization. To visualize their statistics, e.g.,
using uncertainty visualization techniques, can be a solution, but only if the features
are based on the same physical quantities and can therefore be directly compared.
New approaches are needed if the underlying multi-field represents a multitude of
physical quantities, in which case features having different meanings are to be com-
bined in one visualization. Extending other feature concepts, such as geometric or
topological ones, to multi-fields will be an additional challenge.
Eugene Zhang on Tensor Fields and their Derived Fields:
Given a tensor field of some order, it is possible to derive a number of tensor fields
from it. Examples of this includes the spatial gradient, the Laplacian, and the diver-
gence. The derived fields contain rich information and provide great insight to the
original field. However, the derived fields often are of a different order. This leads
to the need of simultaneous analysis and visualization of multiple tensor fields of
different types. Most existing work on multi-field analysis focuses on fields of the
same type, and there has not been much research on higher-order tensor fields due
to the mathematical and physics background it often requires.
References
1. Chen, C.: Top 10 unsolved information visualization problems. IEEE Comput. Graph. Appl.
25 (4), 12-16 (2005)
2. Hauser, H., Bremer, P.T., Theisel, H., Trener, M., Tricoche, X.: Panel: What are the most
demanding and critical problems, and what are the most promising research directions in
Topology-Based Flow Visualization? In Topology-Based Methods in Visualization Workshop.
Budmerice, Slovakia (2005)
3. House, D., Interrante, V., Laidlaw, D., Taylor, R., Ware, C.: Panel: Design and evaluation in
visualization research. Proc. IEEE Vis. 2005 , 705-708 (2005)
4. Johnson, C.R.: Top scientific visualization research problems. IEEE Comput. Graph. Appl.
24 (4), 13-17 (2004)
5. Johnson, C.R., Moorehead, R., Munzner, T., Pfister, H., Rheingans, P., Yoo, T. S.: NIH/NSF
Visualization Research Challenges (Final Draft, Jan 2006). Technical report (2006)
6. Johnson, G., Ebert, D., Hansen, C., Kirk, D., Mark, B., Pfister, H.: Panel: the future visualization
platform. Proc. IEEE Vis. 2004 , 569-571 (2004)