Geoscience Reference
In-Depth Information
In our case, the ontology aims at organizing and formally representing
visualization techniques for 3D city models. Existing ontologies cannot be reused
because they mainly focus on the visualization process and not on the required
items for an effective representation and reuse of visualization techniques.
2.3 Evaluation Representation
When designing a 3D city model, the designer faces the problem of choosing the
most relevant visualization techniques for viewing the geometry and the associated
information. This means that the selected techniques must at least be able to dis-
play the desired information (effectivity); efficiently support the user task(s); not
negatively interfere with each other (e.g. by hiding information). Given the vast
amount of visualization techniques that have been developed over the last decades,
and the variety of visual contexts, user tasks, and data types, the selection of rel-
evant visualization techniques is far from trivial. When the evaluation results are
publicly available they can be used to help selecting a technique.
Evaluating a visualization technique for some user tasks and for a specific con-
text refers to the usability of the technique. From an exhaustive review of litera-
ture, Hornbæk (
2006
) defined usability aspects in terms of effectiveness, efficiency
and satisfaction, the usability being the effectiveness, efficiency, and satisfac-
tion with which specified users could achieve goals in particular environments.
Bowmann et al. (
2004
) defined a classification of usability evaluation methods
for virtual environments. He differentiated three criteria: (1) the user involvement
(whether the evaluation requires or does not require users) (2) the context of evalu-
ation (the context being either generic or application specific) and (3) the type of
results (quantitative or qualitative). By combination, those criteria define height
types of evaluation methods.
Several authors have defined a taxonomy of tasks in the field of information
visualization. Amar et al. (
2005
) have defined low-level tasks that capture the
activities of people trying to use information visualization tools for understand-
ing data. Lee et al. (
2006
) have defined a taxonomy for viewing graphs. In this
approach, complex tasks are considered as a series of low-level tasks.
In order to access information in 3D virtual environments, a user has to perform
manipulation and locomotion. Tyndiuk (
2005
) defined six different types of spatial
configuration from two reference frames: the user's viewpoint reference frame and
the movement reference frame. According to the first one the user can be inside or
outside the scene while the second one refers to the movement whether it derives
from the camera, from the object or from the scene itself. Thereby, whether the
user is inside or outside the scene, he/she can move, manipulate objects or manip-
ulate the scene itself. For 3D urban models manipulation and/or locomotion is
most often done according to a viewpoint close to the ground (snail view or pedes-
trian level) or by flying through or over the model (bird's eye view) (Vaaraniemi
et al.
2013
).
