Information Technology Reference
In-Depth Information
Early researchers were aware that attributes such as colour and shape
are multi-dimensional and in some cases provide rich multi-attribute
examples e.g. [Ber67]; but most early researchers indicate shape as an
attribute without much elaboration and typically refer to the global form,
i.e. simple regular geometric primitives such as circles, squares, kites,
stars, teardrops, etc.
Different researchers can be ambiguous with terminology. For
example, the terms angle, slope, orientation, and collinearity refer to
orientation of different elements in a scene. In this chapter, orientation
indicates a rotational transformation upon an entire glyph within a plot,
while angle indicates the interior angle formed by edges within a singular
glyph. As shown in the table above and in the figure below (Fig 3.6),
orientation and angle should be considered two separate, independent
visual attributes.
Fig. 3.6.
Both orientation and angle can encode information. In the left image, the
only difference is the orientation of an entire single glyph. In the right image, the
only difference is the interior angle of a single glyph.
Early information visualization researchers limited the effectiveness of
shape attributes to only categorical visual tasks (with the exception of
Mazza on curvature). Pre-attentive perception researchers limit their
findings to binary tasks - either the existence of the visual attribute or not.
As shown in later examples, shape can be used to encode either categorical
data or quantitative data.
Scientific visualization
The field of scientific visualization has used shapes - particularly
curvature - for several decades as a means to convey data within a glyph.
Most of the techniques have been focused on generating smooth, curved
shapes to represent continuous quantitative data, such as tensor data. For
example (Fig. 3.7):
Search WWH ::
Custom Search