Geoscience Reference
In-Depth Information
a system that can handle a large number of techniques and incorporate new tech-
niques without explicitly programming (coding) each technique. This means that
the formal description of a technique must be sufficient for the system to instanti-
ate it in a scene for a given dataset.
6.1 Approach
The chosen approach consists in defining a visualization generation environment
that has to be validated on test cases including several 3D scenes, several datasets
and several technique descriptions.
One can view the visualization generation as a process that adds new visual
objects to the given 3D model or modifies objects of the model.
A technique description is essentially a mapping that associates input data ele-
ments to output objects. For instance, if input data are triples (
x, y, t
) that represent
a temperature
t
measured at point (
x, y
), a “cone” representation technique could
map each triple (
x, y, t
) to a cone with its base centre at (
x, y
) and a height of
t
.
More sophisticated techniques would of course require more complex mappings.
6.2 Specification of the Environment
From a high level point of view the environment should support the following
lifecycle:
1. The user selects a data set and a 3D city model.
2. If needed, the user defines a mapping between data and objects in the 3D city
model (e.g. energy data related to buildings).
3. The user selects a visualization technique among the available techniques.
4. The system generates an enriched 3D view that contains the 3D city model and
the data representation according to the selected technique.
5. The user performs tasks in the generated 3D view (or asks testers to perform
the tasks and measures the technique's usability).
6. If necessary go back to step 3.
7 Conclusion
In this chapter we address the general problem of finding usable information visu-
alization techniques in 3D city models. We propose (1) a model of such techniques
and thus a common detailed and organized description of 3D visualization tech-
niques that are too often not explicitly described even when they are widely used
(2) a formalized implementation of this model in the form of an ontology. The
