Spatial Cues in 3D Visualization - Ambient Intelligence for Scientific Discovery - page 107

than” overall effect. Thus, instead of multiple cues incrementally adding more
information about depth, the multiple cues interact, and may, in the aggregate,
provide more, or less, perceptual information about the depth of objects seen
within our visual fields.
2.2 Key Depth Cue Concepts
Motion. The typical human perceptual experience occurs within a context of
nested motions. The movement of the eyes, as well as the movement of objects,
provide important perceptual cues about the environmental and spatial proper-
ties of objects perceived in space.
Motion ,or motion parallax , has previously been demonstrated to be a pow-
erful depth cue [6-11]. However, there is conflicting evidence whether user-
controlled object motion provides more depth information than uncontrolled
motion. In [12], they reported no differences in performance due to controlled
and uncontrolled motion, whereas [8] reported improved performances with user-
controlled motion. Furthermore, there is evidence that it does not matter
whether the object or the observer is moving [13]. Research has indicated that
the motion cue can be introduced simply through the observer's own head move-
ment [14, 15]. For example, [16] and [17] reported that introducing the motion
cue through the use of head-tracking displays had performance effects as pow-
erful as stereo viewing.
Stereopsis. Stereopsis, or the perception of stereoscopic images, is enabled
by the physiological condition of binocular disparity. Because human eyes are
positioned approximately six centimeters apart, an environmental object that
is viewed with both eyes within a distance of about thirty meters projects two
separate and displaced images onto the retinas of the left and right eyes. The
brain fuses these two images, and in the process, the viewer is provided with
significant depth cues regarding the relative size, shape, orientation and distance
of the viewed object. However, we nevertheless perceive depth, and can make
reasonable estimates about the relative sizes and locations of even the most
distant objects viewed in our environment. Thus, stereoscopic viewing is not the
only effective visual mechanism for perceiving depth.
Human performance in a variety of task domains using stereoscopic user
interfaces has been previously investigated. Relevant domains and tasks have
included the viewing, manipulation, and/or recognition of object images [7, 8,
12, 18-22]; relative depth perception [23]; medicine [9]; and cockpit situational
awareness [24-26]. Stereo viewing has been demonstrated to be a powerful vi-
sual cue for understanding depth information [27, 28], and is sometimes used
as a baseline condition for assessing the relative ecacies of other 2D and 3D
visualization techniques [8, 16, 19, 20].
Studies have demonstrated task performance advantages from stereo viewing,
particularly in perceiving, recognizing, grasping, moving, positioning and resiz-
ing objects viewed in depth, as well as in recognizing and understanding object
shapes [8, 9, 12, 18-22, 29]. Nevertheless, it is not universally accepted that stereo-
scopic viewing is a predominant depth cue [7]. Some researchers have suggested