ArticlePDF Available

Abstract

Natural walking can provide a compelling experience in immersive virtual environments, but it remains an implementation challenge due to the physical space constraints imposed on the size of the virtual world. The use of redirection techniques is a promising approach that relaxes the space requirements of natural walking by manipulating the user's route in the virtual environment, causing the real world path to remain within the boundaries of the physical workspace. In this paper, we present and apply a novel taxonomy that separates redirection techniques according to their geometric flexibility versus the likelihood that they will be noticed by users. Additionally, we conducted a user study of three reorientation techniques, which confirmed that participants were less likely to experience a break in presence when reoriented using the techniques classified as subtle in our taxonomy. Our results also suggest that reorientation with change blindness illusions may give the impression of exploring a more expansive environment than continuous rotation techniques, but at the cost of negatively impacting spatial knowledge acquisition.
A Taxonomy for Deploying Redirection
Techniques in Immersive Virtual Environments
Evan A. Suma
Gerd Bruder
Frank Steinicke
David M. Krum
Mark Bolas
USC Institute for Creative Technologies
University of W
¨
urzburg
ABSTRACT
Natural walking can provide a compelling experience in immersive
virtual environments, but it remains an implementation challenge
due to the physical space constraints imposed on the size of the
virtual world. The use of redirection techniques is a promising
approach that relaxes the space requirements of natural walking by
manipulating the user’s route in the virtual environment, causing
the real world path to remain within the boundaries of the physical
workspace. In this paper, we present and apply a novel taxonomy
that separates redirection techniques according to their geometric
flexibility versus the likelihood that they will be noticed by users.
Additionally, we conducted a user study of three reorientation
techniques, which confirmed that participants were less likely
to experience a break in presence when reoriented using the
techniques classified as subtle in our taxonomy. Our results also
suggest that reorientation with change blindness illusions may
give the impression of exploring a more expansive environment
than continuous rotation techniques, but at the cost of negatively
impacting spatial knowledge acquisition.
Keywords: Virtual environments, redirection, taxonomy
Index Terms: H.5.1 [[Information Interfaces and Presenta-
tion]: Multimedia Information Systems—Artificial, augmented,
and virtual realities; I.3.6 [Computer Graphics]: Methodology and
Techniques—Interaction techniques; I.3.7 [Computer Graphics]:
Three-Dimensional Graphics and Realism—Virtual reality
1 I
NTRODUCTION
Natural interaction is vitally important for creating compelling vir-
tual reality experiences, particularly locomotion, which is one of
the most common and universal tasks performed when interacting
with 3D graphical environments [1]. The most natural locomotion
technique, real walking, has been shown to provide a greater sense
of presence when compared to alternative techniques that do not
employ realistic body motion, including walking-in-place and vir-
tual travel metaphors (e.g. flying) [18], and has also been shown to
provide benefits for memory and attention [16]. Despite these ad-
vantages, natural walking remains a challenge for practitioners that
utilize immersive head-mounted displays, as physical space limita-
tions will ultimately restrict the size of the virtual environment that
can be explored.
Redirection is a promising solution that relaxes the space restric-
tions of natural walking by manipulating the user’s route in the vir-
tual environment, causing it to deviate from the real world path [13].
These techniques can be used to allow considerably larger virtual
environments to be explored using natural body motions within a
relatively confined physical workspace. To better understand how
e-mail: {suma, krum, bolas}@ict.usc.edu
e-mail: {gerd.bruder, frank.steinicke}@uni-wuerzburg.de
Taxonomy of Redirection Techniques
Overt Subtle
Reorientation (q)
Repositioning (XYZ)
Discrete
freeze-and-turn
resetting
[20]
Discrete
change blindness
architectural illusions
[17]
Continuous
rotation or
curvature gains
[5] [10] [13] [15]
Continuous
rotation gains
with interventions
[11] [12] [20]
Continuous
gradual translation
(e.g. escalators)
[6]
Discrete
teleportation
[2] [14]
Discrete
change blindness
self-motion Illusions
[4]
Continuous
translation gains
[8] [19]
Figure 1: Taxonomy of redirection techniques for supporting natural
walking through immersive virtual environments. The vertical axis
distinguishes how the technique is applied in the environment. The
horizontal axis provides a ranking in terms of noticeability to the user.
The division in cells represents distinct implementation strategies for
each type of technique.
these redirection techniques may be employed in practice, it is use-
ful to provide a classification scheme that maps the spectrum of
available methods. Steinicke et al. previously described one such
taxonomy for redirection techniques that manipulate locomotion
gains, based on the type of gain being applied (translation, rotation,
or curvature) [15]. However, recent work has yielded an assortment
of innovative redirection techniques with similar goals, but drasti-
cally different implementations that do not neatly fit into previous
conceptual frameworks. As such, we have developed a new taxon-
omy that is organized to allow practitioners to select and apply one
or more redirection metaphors based on a variety of criteria relevant
to virtual environment design. Additionally, we conducted a user
study of three techniques selected from the taxonomy and collected
data on participants’ breaks in presence and spatial knowledge ac-
quisition during redirection.
2 T
AXONOMY / PREVIOUS WORK
Figure 1 illustrates the taxonomy, which is based on each redirec-
tion technique’s geometric applicability, noticeability to the user,
and content-specific implementation details. Redirection tech-
43
IEEE Virtual Reality 2012
4-8 March, Orange County, CA, USA
978-1-4673-1246-2/12/$31.00 ©2012 IEEE
niques may be broadly divided into two categories based on their
geometric applicability, in other words, how they help make the
desired virtual space fit within the actual tracked volume. Reposi-
tioning techniques manipulate the correspondence between points
in the real and virtual worlds to compress a larger virtual space
into a smaller physical workspace. Reorientation techniques at-
tempt to rotate the user’s heading away from the boundaries of the
physical workspace. Ideally, the user would not notice redirection
techniques, so that the virtual reality implementation remains as in-
visible as possible. As such, our proposed taxonomy distinguishes
between subtle and overt methods. Subtle techniques are designed
specifically to be imperceptible when the magnitude of the manip-
ulation is beneath a certain detection threshold. In contrast, overt
techniques will be easily noticed by users when they are applied.
While the subtlety of techniques can most likely be mapped to a
continuum, this course binary categorization is nonetheless concep-
tually useful for our taxonomy. This distinction implies a logical
ranking: in general, subtle techniques are preferable to overt ones,
since the latter have greater potential to disrupt the natural process
of walking through the environment. Finally, in terms of implemen-
tation, techniques may either be discrete (applied instantaneously)
or continuous (applied over time), which is a potentially important
factor as practioners seek to balance the implementation of these
techniques with the narritive and timing demands of their content.
2.1 Repositioning Techniques
Overt Continuous Repositioning. A simple repositioning can be
achieved by continuously translating the virtual environment about
the user’s position. This allows the user to walk to areas in the
virtual environment that were not previously accessible within the
confines of the physical workspace. This may be disorienting if the
virtual world is translated unexpectedly, and may make the virtual
environment appear unstable. This disruption can be mitigated by
coupling the translation with known metaphors associated with mo-
tion, such as elevators (e.g. [7]), escalators (e.g. as demonstrated
with [6]), moving walkways, or vehicles.
Subtle Continuous Repositioning. A continuous repositioning
can be applied in a subtle manner by applying translation gains
to the user’s physical locomotion, effectively scaling walking mo-
tions to cover greater distances in the virtual environment [19]. This
method can be improved by estimating the user’s intended direction
of travel and scaling translations only in that direction, which re-
duces exaggeration of the oscillatory head bob and sway from walk-
ing motions [8]. This technique remains subtle so long as the gains
applied are small enough to avoid detection. Steinicke et al. con-
ducted a psychophysical study of detection thresholds, and found
that travel distances could be downscaled by 14% or upscaled by
26% without becoming noticeable to the user [15].
Overt Discrete Repositioning. Discrete repositioning techniques
may be achieved through instantaneous translation, effectively tele-
porting the user to a new location in the virtual space. This tech-
nique is potentially disorienting if the user is not expecting the
virtual position to be manipulated. To mitigate this problem, re-
searchers have leveraged the concept of portals from popular sci-
ence fiction to provide an environmental grounding for teleportation
[2]. In fact, users have reported greater levels of presence when por-
tals were used to teleport from a transitional virtual replica of the
real environment into an unfamiliar environment, compared to en-
tering the unfamiliar environment immediately [14].
Subtle Discrete Repositioning. Given the abrupt translation re-
quired for a discrete repositioning, it seems difficult to apply this
technique in a subtle manner. However, recent research has found
that inter-stimulus images or visual optic flow effects in the periph-
ery of the user’s view can be used to mask abrupt translations in the
environment [4]. These small discrete updates can be repeated pe-
riodically as the user walks, allowing travel distances to be scaled
similar to the continuous techniques.
2.2 Reorientation Techniques
Overt Continuous Reorientation. Resetting is a conceptually simple
method of reorienting the user that requires an intervention when
the user reaches the boundaries of the physical workspace. During
the intervention, the user is instructed to turn around, during which
a rotation gain is applied. For example, for every one degree of
turn in the real world, the virtual environment is rotated two de-
grees, so that after a 180 degree physical turn (pointing back into
the workspace) the virtual environment has rotated by 360 degrees,
restoring the user’s original heading in the virtual world [20]. Since
issuing explicit instructions to the user may break presence, one
suggested mitigator has been the use of visual distractors to elicit
head turns during the intervention, thereby providing an opportu-
nity to apply rotation gains [11]. It is important to note that while
the continuous rotation gain itself may not be detectable, interven-
tions are obvious to users, and so we classify the overall method as
an overt technique.
Subtle Continuous Reorientation. Redirected walking was the
first technique to introduce an imperceptible gain to head rotations
in order to guide the user away from the boundaries of the physi-
cal workspace [13]. The detectability of rotation gains have been
studied in the context of both head turns [9] and full-body turns
[3]. A recent comprehensive study found that users can be phys-
ically turned approximately 49% more or 20% less than the per-
ceived virtual rotation without noticing [15]. Alternatively, it is
also possible to dynamically apply a continuous rotation as the user
travels forwards, resulting in a curvature of the walking path [5]
[13]. However, experimental results have indicated that impercep-
tibly redirecting a user along a circular arc requires a very large
workspace with a radius of at least 22 meters [15].
Overt Discrete Reorientation. In addition to continuous tech-
niques, resetting can also be applied in a discrete manner. In their
freeze-and-turn implementation of resetting, Williams et al. freeze
motion tracking and instruct the user to rotate away from the bound-
aries of the physical workspace [20]. After the user completes the
physical rotation, the virtual view is unfrozen and motion tracking
resumes. While the discontinuity introduced by freezing and un-
freezing the motion tracking will be obvious to the user, resetting
remains useful as an emergency “failsafe” technique to prevent the
user from exiting the workspace.
Subtle Discrete Reorientation. In a drastically different imple-
mentation from other reorientation techniques, researchers have
proposed instantaneously changing the location or orientation of ar-
chitectural features, particularly doors, in a virtual scene at runtime
[17]. The technique is based on change blindness, a phenomenon
that can be observed when users fail to notice alterations to a visual
scene that occur outside of their visual field. While studies have
shown that this illusion can be leveraged to reorient the user in a
very subtle way, with only one out of 77 users noticing the scene
change, change blindness techniques are largely limited to interior
environments with doorways that can be manipulated, and would
often not be geometrically applicable in sparse, open environments
such as outdoor scenes.
2.3 Redirection Controllers
Each redirection technique imposes its own set of limitations, mak-
ing it difficult or impossible to provide unlimited free exploration
with a single technique. Therefore, to employ redirection in prac-
tical virtual environments, redirection controllers must maintain
awareness of the user’s state in real and virtual space, and invoke
repositioning and/or reorientation techniques in order to facilitate
walking through the virtual world. A fairly restrictive example is a
waypoint-based controller, which reorients users as they walk be-
tween predefined locations, usually along a zig-zag or “S” curve
44
path (e.g. [13]). More sophisticated controllers have attempted to
dynamically optimize continuous reorientation techniques, for ex-
ample, by adjusting curvature gain levels based on the user’s walk-
ing speed [10]. Redirected Free Exploration with Distractors is per-
haps the most generally applicable redirection controller developed
to date, which continuously steers the user towards the center of the
tracked area and invokes overt continuous reorientation when the
user approaches the boundaries of the physical space [12].
Developing automated redirection controllers that utilize a wider
range of available techniques will be an important step towards
making redirection more generally applicable for practical settings.
One example of an implementation that makes use of both discrete
and continuous techniques is Arch-Explore, a system for architec-
tural walkthroughs that combines repositioning using discrete por-
tals and continuous translation, rotation, and curvature gains for
reorientation in a semi-automated manner [2]. We believe the tax-
onomy presented in this paper will provide a useful starting point
for the design of future redirection controllers that make use of mul-
tiple repositioning and reorientation techniques in tandem.
3 U
SER STUDY
Based on the taxonomy presented in Section 2, we selected three
existing reorientation techniques and asked participants to self-
report whenever they experienced a break in presence. We hypoth-
esized that continuously collecting data on these breaks in presence
as participants were redirected would be a useful metric for mea-
suring the “subtlety” or “overtness” of each technique.
3.1 Study Design
We chose to focus on reorientation techniques since they are more
commonly used and cited in the literature than repositioning tech-
niques, and tested one approach from each of the three parts of the
taxonomy that were most used in practice, i. e., excluding overt dis-
crete manipulations. We conducted a within-subjects study with all
participants experiencing the following three conditions:
SCR: Subtle Continuous Reorientation
To implement this technique, we applied rotation gains as users
walked around a partially-opened virtual swing door connecting
two virtual rooms, as proposed by [2]. The detectability of these
manipulations depends mainly on the discrepancy between a ma-
nipulated virtual rotation
α
virtual
compared to a rotation of a user
in the real world
α
real
, expressed via rotation gains:
α
virtual
=
g
R
·
α
real
, for g
R
R [15]. In this notation, g
R
= 1 would imply an
exact 1:1 mapping from real to virtual rotation; therefore, manip-
ulations become less noticeable as g
R
approaches 1. To compare
this technique in both optimal and non-optimal cases, we compared
results between two different situations: reorientations achievable
with g
R
>= 0.59 and reorientations requiring g
R
< 0.59, (cf. [2]).
SDR: Subtle Discrete Reorientation
We implemented the change blindness reorientation technique to
manipulate the location of virtual doorways, as was done in [17].
To compare this technique in both optimal and non-optimal cases,
we conducted an informal pilot test to determine the magnitude of
scene changes that can be feasibly applied. Thus, we compared
results between two different situations: reorientations achievable
with door movement distances <= 1m and reorientations requiring
distances > 1m.
OCR: Overt Continuous Reorientation
We implemented rotation gains with distractors using an animated
virtual hummingbird and a gain of 1.5 of the user’s head rotation,
as suggested by Peck et al. [11]. Since this technique is overt and
cannot be applied without the user noticing, it did not make sense
to discriminate between optimal and non-optimal cases.
3.2 Methods
A total of 22 people participated in the experiment (16 male, 6
female). Participants were university students between the ages
of 2130 (M = 24.4), and had normal or corrected-to-normal vi-
sion. The graphical environment was presented on a ProView SR80
HMD manufactured by Kaiser Electro-Optics (1280 × 1024 resolu-
tion, 60Hz refresh rate, 80
diagonal field of view) with an opaque
cloth attached to block peripheral vision of the real world. The po-
sition of the HMD was tracked with an infrared LED and an active
optical tracking system (Precision Position Tracker PPT X8 from
WorldViz), which provides sub-millimeter precision and an update
rate of 60Hz. Orientation tracking was achieved with an InterSense
InertiaCube 3 fixed to the top of the HMD. The virtual environment
was rendered at 60 frames per second using OpenGL on PC with In-
tel Core i7 processors, 6GB of RAM, and nVidia Quadro FX 4800
graphics card.
At the beginning of each session, participants were guided into
the laboratory room wearing a blindfold to avoid exposing them to
the physical workspace. They were instructed to explore a virtual
environment consisting of a series of offices arranged in a randomly
generated layout. The experimental task required participants to
collect a one dollar bill from an avatar in each room, then proceed
towards an adjacent office via a color-identified door. Avatars were
matched to a student coworker in the laboratory, who assumed the
corresponding pose to provide passive haptic feedback. Participants
were instructed to announce verbally whenever they experienced a
break in presence (BIP), which we described to them as the feeling
that the virtual scene or interaction appeared implausible. After the
VR session, subjects sketched the path they traveled through the
VE by drawing a virtual floor plan on a sheet of blank paper, ex-
cluding furniture or avatars. The maps were evaluated separately
for each transition between rooms, to which a score of either +1
was assigned if the path information between entering and leaving
the room roughly matched the actual door layout (i. e., the unmodi-
fied door layout in the case of SDR), or 0 otherwise. The total map
score was computed as sum of the scores for the separate transitions
for each subject, and varied between 010. Furthermore, we asked
subjects to label the sides of a square map with their estimation of
the dimensions of the physical walking area in the laboratory. The
total time to complete the study was approximately one hour.
3.3 Results
Figure 2 shows the pooled BIP probabilities in each of the five con-
ditions. The results were treated with a repeated measures analysis
of variance (ANOVA) with a significance level of
α = .05, which
was significant, F(4,84) = 76.43, p < .01,
η
2
p
= .78. Pairwise com-
parisons with Bonferroni-adjusted
α values indicated that SCR op-
timal (M = .13, SD = .13) and SDR optimal (M = .21, SD = .28) had
lower BIP probabilities than all other conditions, p < .01, but were
not significantly different from each other, p > .99. Additionally,
OCR (M = .91, SD = .14) had higher BIP probabilities than SCR
non-optimal (M = .70, SD = .18), p < .01, but was not significantly
different from SDR non-optimal (M = .86, SD = .19), p > .99.
Repeated measures ANOVAs testing the within-subjects effect
of reorientation technique were performed for both the sketch map
grades and physical room size estimates. Significant results were
observed for the map ratings, F(2,42) = 17.26, p < .01,
η
2
p
= .45.
Pairwise comparisons indicated that sketch maps were rated lower
in the SDR condition (M = 6.68, SD = 2.08) compared to both SCR
(M = 8.95, SD = 1.89), p < .01, and OCR (M = 8.91, SD = 1.48),
p < .01. The ratings were not significantly different between SCR
and OCR, p > .99. The results for the room size estimations were
also significant, F(2,42) = 88.29, p < .01,
η
2
p
= .81. Pairwise com-
parisons indicated that the length of a wall in the square physical
room were estimated to be the longest when reorienting with SDR
(M = 12.43m, SD = 3.36m), compared to the shorter estimates in
45
OCRSDR (non-
optimal)
SCR (non-
optimal)
SDR
(optimal)
SCR
(optimal)
Probability of BIP
1.00
0.80
0.60
0.40
0.20
0.00
0.21
0.13
0.91
0.86
0.70
Error Bars: 95% CI
Figure 2: Results showing the pooled probability of a reported a
break in presence for each condition. Subjects reported significantly
fewer breaks during reorientation using the SCR and SDR tech-
niques when they were applied optimally.
both the SCR (M = 6.72m, SD = 1.38m), p < .01, and OCR condi-
tions (M = 5.05, SD = 0.74m), p < .01. The difference between the
estimates in the SCR and OCR conditions was significant, p < .01.
3.4 Discussion
Self-reported breaks in presence seem to be a useful metric for dis-
criminating between subtle and overt redirection techniques. The
results from the study confirmed that reorienting the user with sub-
tle techniques (SCR and SDR) is preferable, but primarily when
they can be applied optimally. However, even when this is not pos-
sible, it may still be beneficial to employ a technique such as SCR
in non-optimal conditions. This is supported by the observation that
even though the SCR non-optimal caused a fairly high incidence of
self-reported BIPs, it was still lower than the OCR condition. How-
ever, overt techniques generally have the advantage of being more
generally applicable, and so any automated redirection controller
should include at least one to prevent failure cases, when the user
would otherwise exit the physical workspace.
On average, subjects estimated the physical workspace to be
much larger when exploring the environment in the SDR condi-
tion, compared to both other conditions as well as th. This interest-
ing result suggests that change blindness architectural illusions may
be more effective at giving the impression of exploring an expan-
sive environment. However, this advantage is not without a cost,
since subjects also received lower sketch map grades compared to
the other two techniques. This is not particularly surprising, since
the environment model was dynamically changing, and the sketch
maps grades were calculated as compared to the original, unmanip-
ulated environment layout. This suggests that these architectural
illusions may not be appropriate for practical use in applications
where acquiring accurate spatial knowledge is important.
4 C
ONCLUSION
In this paper, we introduced a novel taxonomy that maps the spec-
trum of available redirection techniques. In the future, this taxon-
omy may be used to inform the design of virtual environments, and
provides the theoretical foundation for the development of auto-
mated redirection controllers that can dynamically apply a variety
of techniques based upon the needs of the system and the current
state of the user.
R
EFERENCES
[1] D. A. Bowman, E. Kruijff, J. J. LaViola, and I. Poupyrev. 3D User In-
terfaces: Theory and Practice. Addison Wesley Longman Publishing
Co., Inc., Redwood City, CA, USA, 2004.
[2] G. Bruder, F. Steinicke, and K. H. Hinrichs. Arch-explore: A natu-
ral user interface for immersive architectural walkthroughs. In IEEE
Symposium on 3D User Interfaces, pages 75–82, 2009.
[3] G. Bruder, F. Steinicke, K. H. Hinrichs, and M. Lappe. Reorientation
during body turns. In Joint Virtual Reality Conference of EGVE - ICAT
- EuroVR, pages 145–152, 2009.
[4] G. Bruder, F. Steinicke, and P. Wieland. Self-motion illusions in im-
mersive virtual reality environments. In IEEE Virtual Reality, pages
39–46, 2011.
[5] D. Engel, C. Curio, L. Tcheang, B. Mohler, and H. H. B
¨
ulthoff. A psy-
chophysically calibrated controller for navigating through large envi-
ronments in a limited free-walking space. In ACM Virtual Reality
Software & Technology, pages 157–164, 2008.
[6] S. S. Fisher, M. McGreevy, J. Humphries, and W. Robinett. Virtual
environment display system. In workshop on Interactive 3D graphics,
pages 77–87, 1987.
[7] L. F. Hodges, R. Kooper, T. C. Meyer, B. O. Rothbaum, D. Opdyke,
J. J. de Graaff, J. S. Williford, and M. M. North. Virtual environments
for treating the fear of heights. Computer, 28:27–34, 1995.
[8] V. Interrante, B. Ries, and L. Anderson. Seven league boots: A new
metaphor for augmented locomotion through moderately large scale
immersive virtual environments. In IEEE Symposium on 3D User In-
terfaces, pages 167–170, 2007.
[9] J. Jerald, T. Peck, F. Steinicke, and M. Whitton. Sensitivity to scene
motion for phases of head yaws. In Symposium on Applied Perception
in Graphics and Visualization, pages 155–162, 2008.
[10] C. T. Neth, J. L. Souman, D. Engel, U. Kloos, H. H. B
¨
ulthoff, and B. J.
Mohler. Velocity-dependent dynamic curvature gain for redirected
walking. In IEEE Virtual Reality, pages 151–158, 2011.
[11] T. C. Peck, H. Fuchs, and M. C. Whitton. Evaluation of reorienta-
tion techniques and distractors for walking in large virtual environ-
ments. IEEE Transactions on Visualization and Computer Graphics,
15(3):383–394, 2009.
[12] T. C. Peck, H. Fuchs, and M. C. Whitton. An evaluation of naviga-
tional ability comparing redirected free exploration with distractors to
walking-in-place and joystick locomotion interfaces. In IEEE Virtual
Reality, pages 55–62, 2011.
[13] S. Razzaque, Z. Kohn, and M. C. Whitton. Redirected walking. In
Eurographics (Short Presentation), 2001.
[14] F. Steinicke, G. Bruder, K. H. Hinrichs, and A. Steed. Gradual transi-
tions and their effects on presence and distance estimation. Computers
& Graphics, 34(1):26–33, 2010.
[15] F. Steinicke, G. Bruder, J. Jerald, H. Frenz, and M. Lappe. Estima-
tion of detection thresholds for redirected walking techniques. IEEE
Transactions on Visualization and Computer Graphics, 16(1):17–27,
2010.
[16] E. Suma, S. Finkelstein, M. Reid, S. Babu, A. Ulinski, and L. Hodges.
Evaluation of the cognitive effects of travel technique in complex real
and virtual environments. IEEE Transactions on Visualization and
Computer Graphics, 16:690–702, 2010.
[17] E. A. Suma, S. Clark, S. L. Finkelstein, Z. Wartell, D. M. Krum, and
M. Bolas. Leveraging change blindness for redirection in virtual en-
vironments. In IEEE Virtual Reality, pages 159–166, 2011.
[18] M. Usoh, K. Arthur, M. C. Whitton, R. Bastos, A. Steed, M. Slater,
and F. P. Brooks. Walking > walking-in-place > flying, in virtual
environments. In ACM Conference on Computer Graphics and Inter-
active Techniques (SIGGRAPH), pages 359–364, 1999.
[19] B. Williams, G. Narasimham, T. P. McNamara, T. H. Carr, J. J. Rieser,
and B. Bodenheimer. Updating orientation in large virtual environ-
ments using scaled translational gain. In Symposium on Applied Per-
ception in Graphics and Visualization, pages 21–28, 2006.
[20] B. Williams, G. Narasimham, B. Rump, T. P. McNamara, T. H. Carr,
J. Rieser, and B. Bodenheimer. Exploring large virtual environments
with an HMD when physical space is limited. In Symposium on Ap-
plied Perception in Graphics and Visualization, pages 41–48, 2007.
46
... Another recent work by Williams et al. [174] introduced alignmentbased redirections that minimize collisions with the physical environment. For an in-depth review on redirected walking, we point to the work by Nilsson et al. [103] and Suma et al. [151]. Instead of imperceptibly changing the user's position, other concepts alter the environment overtly to achieve similar e ects [134]. ...
Thesis
Once a topic only for researchers and enthusiasts, virtual reality (VR) has recently developed into a widely available platform with huge potential. However, we are still far from tapping the full potential of virtual environments. Whereas one might argue that the reasons reside in the low prevalence of headsets or the necessity for further technical advancements, we see a primary reason in the expectations for VR. Often, it is tempting to copy tried-and-tested interactions and interfaces from non-VR applications or replace established approaches and workflows that work well without a VR headset. Instead, we want to think outside the box and design techniques "VR-first" that leverage the unique advantages VR oers. In this dissertation, we explore how to design the interaction with virtual worlds to achieve a natural and fluent VR experience. Our work spans four essential aspects of VR research: locomotion, interaction, perspectives, and applications. First, we contribute to the field of locomotion research by establishing four unique navigation concepts that either target decisive gaps in the literature or improve existing approaches. Next, we focus on the interaction within the VR environment by presenting our eorts in un- derstanding user behavior, imagining novel input modalities, and structuring interface design. Afterward, we extend the sensation of owning a virtual body in VR to animal avatars and investigate the potential of multiprotagonist narratives where users switch between dierent characters. In the last part, we cover the general use of VR in every- day life. Therefore, we explore how non-VR audiences can watch the user’s experience in the virtual world. We also demonstrate the potential of full-body exercises in VR by designing an exergame for safe and engaging jump training. Finally, we conclude the dissertation with a general discussion of the presented concepts and a critical look at our research and its potential impact.
... Subtle resemblances of diegetic methods work on an unconscious level, while overt is non-diegetic and noticed more easily by the user. We recognize subtle techniques as more appropriate for capturing attention in 360 • video and VR systems, since they are less likely to distract the natural process of watching or interacting with the environment than overt methods [28]. In some cases, it is hard to distinguish between subtle and overt methods, especially in the effects of some post-production techniques and corrections (e.g., saturation, exposure, hue), where the categorization depends on the degree of modification [25]. ...
Article
Full-text available
Sensing and remembering features in visual scenes are conditioned by visual attention and methods to guide it. This should be relevant in terms of product placement, which has become an important part of incorporating brands into different mass media formats with a commercial purpose. The approach can be challenging in 360° video, where an omnidirectional view enables consumers to choose different viewing perspectives, which may result in overlooking the brands. Accordingly, attention guidance methods should be applied. This study is the first to explore diegetic guidance methods as the only appropriate guiding method for an unobtrusive and unconscious nature of product placement. To test the effectiveness of three different diegetic guiding methods, a between-subject design was employed, where the participants were assigned randomly to one of four videos with the same scene but different guiding methods. The findings show and explain the discrepancy with studies on guiding attention in other contexts, as there were no significant differences between the guiding cues according to brand recall and brand recognition. The results also indicate a significant influence of brand familiarity on brand recall in 360° video. The article concludes by providing limitations, future research directions, and recommendations for audiovisual policy.
... La literatura científica relacionada específicamente con los vídeos de 360° narrativos de ficción surgió principalmente a partir de la popularización de este nuevo formato en el año 2015 y, se podría agrupar en cuatro áreas temáticas: estudios sobre cómo guiar la atención del espectador en entornos inmersivos (Suma et al., 2012;Lelyveld, 2015;Nielsen et al., 2016;Brillhart, 2016;Mateer, 2017;Pavel et al., 2017;Gödde et al., 2018;Bender, 2019;Rothe et al., 2019b;Kvisgaard et al., 2019;Schmitz et al, 2020), investigaciones relativas al empleo de elementos del lenguaje audiovisual tradicional que se eliminan o modifican (Ryan, 2015;Dolan y Parets, 2016;Kjaer et al., 2017;Erkut, 2017;Serrano et al., 2017;Nicolae, 2018;Marañes et al., 2020;Masia et al., 2021) y estudios sobre la redefinición del espacio en un filme inmersivo. ...
Article
Los filmes inmersivos son un nuevo formato para contar historias que utiliza el espacio de la imagen como un elemento audiovisual esencial para entender correctamente la narrativa. Hemos realizado un análisis de contenido sobre una muestra compuesta por vídeos de 3600 narrativos de ficción, a la que aplicamos un modelo compuesto por variables relacionadas con la organización del espacio alrededor del espectador. Los resultados señalan un esquema organizativo centrípeto, similar al teatro inmersivo, donde se da prioridad al entendimiento de la narrativa principal por encima de una verdadera utilización del espacio que proporciona este nuevo formato para contar historias.
... Throughout the last few years, numerous improvements [26,31,47] have been proposed, including refining the detection thresholds for different users and conditions [100] or introducing alignment-based redirections that reduce collisions with the physical environment [99]. For an in-depth review of the state of research, we point to the work by Nilsson et al. [64] and Suma et al. [84]. Finally, the Space Bender concept by Simeone et al. [74] achieves a comparable effect by overtly altering the environment instead of the users' movements. ...
Article
The size of most virtual environments exceeds the tracking space available for physical walking. One solution to this disparity is to extend the available walking range by augmenting users' actual movements. However, the resulting increase in visual flow can easily cause cybersickness. Therefore, we present a novel augmented-walking approach for virtual reality games. Our core concept is a virtual tunnel that spans the entire travel distance when viewed from the outside. However, its interior is only a fraction as long, allowing users to cover the distance by real walking. Whereas the tunnel hides the visual flow from the applied movement acceleration, windows on the tunnel's walls still reveal the actual expedited motion. Our evaluation reveals that our approach avoids cybersickness while enhancing physical activity and preserving presence. We finish our paper with a discussion of the design considerations and limitations of our proposed locomotion technique.
Article
Full-text available
Target selection in virtual reality (VR) is usually carried out with the need of visual attention. While target selection in VR has been extensively investigated in non-walking activities (e.g., sitting or standing), there have been few studies about eyes-engaged target selection during walking in virtual environments. Therefore, we conducted a comprehensive study to explore the effects of physical walking (as an independent variable with low, medium and high speeds) on eyes-engaged selection tasks with targets (three target sizes and three target depths) in two experiments: targets fixed in the virtual environment (Experiment One) and targets fixed to the virtual body (Experiment Two), respectively. Results showed that for Experiment One, the low walking speed led to the significantly longest task completion time, while the medium and high speeds had similar task completion time. For Experiment Two, higher walking speed led to longer task completion time. In both tasks, error rate significantly increased as walking speed increased. The effects of walking speed also varied across target size and target depth. We conclude our study with a set of design implications for target selection tasks when walking in VR environments.
Article
In multi-user Redirected Walking (RDW), the space subdivision method divides a shared physical space into sub-spaces and allocates a sub-space to each user. While this approach has the advantage of precluding any collisions between users, the conventional space subdivision method suffers from frequent boundary resets due to the reduction of available space per user. To address this challenge, in this study, we propose a space subdivision method called Optimal Space Partitioning (OSP) that dynamically divides the shared physical space in real-time. By exploiting spatial information of the physical and virtual environment, OSP predicts the movement of users and divides the shared physical space into optimal sub-spaces separated with shutters. Our OSP framework is trained using deep reinforcement learning to allocate optimal sub-space to each user and provide optimal steering. Our experiments demonstrate that OSP provides higher sense of immersion to users by minimizing the total number of reset counts, while preserving the advantage of the existing space subdivision strategy: ensuring better safety to users by completely eliminating the possibility of any collisions between users beforehand. Our project is available at https://github.com/AppleParfait/OSP-Archive.
Article
Full-text available
A head-mounted, wide-angle, stereoscopic display system controlled by operator position, voice and gesture has been developed for use as a multipurpose interface environment. The system provides a multisensory, interactive display environment in which a user can virtually explore a 360-degree synthesized or remotely sensed environment and can viscerally interact with its components. Primary applications of the system are in telerobotics, management of large-scale integrated information systems, and human factors research. System configuration, application scenarios, and research directions are described.
Article
Full-text available
Redirected walking techniques allow people to walk in a larger virtual space than the physical extents of the laboratory. We describe two experiments conducted to investigate human sensitivity to walking on a curved path and to validate a new redirected walking technique. In a psychophysical experiment, we found that sensitivity to walking on a curved path was significantly lower for slower walking speeds (radius of 10 m versus 22 m). In an applied study, we investigated the influence of a velocity-dependent dynamic gain controller and an avatar controller on the average distance that participants were able to freely walk before needing to be reoriented. The mean walked distance was significantly greater in the dynamic gain controller condition, as compared to the static controller (22 m versus 15 m). Our results demonstrate that perceptually motivated dynamic redirected walking techniques, in combination with reorientation techniques, allow for unaided exploration of a large virtual city model.
Conference Paper
Full-text available
We present change blindness redirection, a novel technique for allowing the user to walk through an immersive virtual environment that is considerably larger than the available physical workspace. In contrast to previous redirection techniques, this approach, based on a dynamic environment model, does not introduce any visual-vestibular conflicts from manipulating the mapping between physical and virtual motions, nor does it require breaking presence to stop and explicitly reorient the user. We conducted two user studies to evaluate the effectiveness of the change blindness illusion when exploring a virtual environment that was an order of magnitude larger than the physical walking space. Despite the dynamically changing environment, participants were able to draw coherent sketch maps of the environment structure, and pointing task results indicated that they were able to maintain their spatial orientation within the virtual world. Only one out of 77 participants across both both studies definitively noticed that a scene change had occurred, suggesting that change blindness redirection provides a remarkably compelling illusion. Secondary findings revealed that a wide field-of-view increases pointing accuracy and that experienced gamers reported greater sense of presence than those with little or no experience with 3D video games.
Conference Paper
Full-text available
We report on a user study evaluating Redirected Free Exploration with Distractors (RFED), a large-scale, real-walking, locomotion interface, by comparing it to Walking-in-Place (WIP) and Joystick (JS), two common locomotion interfaces. The between-subjects study compared navigation ability in RFED, WIP, and JS interfaces in VEs that are more than two times the dimensions of the tracked space. The interfaces were evaluated based on navigation and wayfinding metrics and results suggest that participants using RFED were significantly better at navigating and wayfinding through virtual mazes than participants using walking-in-place and joystick interfaces. Participants traveled shorter distances, made fewer wrong turns, pointed to hidden targets more accurately and more quickly, and were able to place and label targets on maps more accurately. Moreover, RFED participants were able to more accurately estimate VE size.
Conference Paper
Full-text available
Motion perception in immersive virtual reality environments significantly differs from the real world. For example, previous work has shown that users tend to underestimate travel distances in immersive virtual environments (VEs). As a solution to this problem, some researchers propose to scale the mapped virtual camera motion relative to the tracked real-world movement of a user until real and virtual motion appear to match, i. e., real-world movements could be mapped with a larger gain to the VE in order to compensate for the underestimation. Although this approach usually results in more accurate self-motion judgments by users, introducing discrepancies between real and virtual motion can become a problem, in particular, due to misalignments of both worlds and distorted space cognition. In this paper we describe a different approach that introduces apparent self-motion illusions by manipulating optic flow fields during movements in VEs. These manipulations can affect self-motion perception in VEs, but omit a quantitative discrepancy between real and virtual motions. We introduce four illusions and show in experiments that optic flow manipulation can significantly affect users' self-motion judgments. Furthermore, we show that with such manipulation of optic flow fields the underestimation of travel distances can be compensated.
Conference Paper
Full-text available
Experience indicates that the sense of presence in a virtual environment is enhanced when the participants are able to actively move through it. When exploring a virtual world by walking, the size of the model is usually limited by the size of the available tracking space. A promising way to overcome these limitations are motion compression techniques, which decouple the position in the real and virtual world by introducing imperceptible visual-proprioceptive conflicts. Such techniques usually precalculate the redirection factors, greatly reducing their robustness. We propose a novel way to determine the instantaneous rotational gains using a controller based on an optimization problem. We present a psychophysical study that measures the sensitivity of visual-proprioceptive conflicts during walking and use this to calibrate a real-time controller. We show the validity of our approach by allowing users to walk through virtual environments vastly larger than the tracking space.
Conference Paper
Full-text available
Virtual Environments presented through head-mounted displays (HMDs) are often explored on foot. Exploration on foot is useful since the afferent and efferent cues of physical locomotion aid spatial awareness. However, the size of the virtual environment that can be explored on foot is limited to the dimensions of the tracking space of the HMD unless other strategies are used. This paper presents a system for exploring a large virtual environment on foot when the size of the physical surroundings is small by leveraging people's natural ability to spatially update. This paper presents three methods of "resetting" users when they reach the physical limits of the HMD tracking system. Resetting involves manipulating the users' location in physical space to move them out of the path of the physical obstruction while maintaining their spatial awareness of the virtual space.
Article
Several experiments have provided evidence that ego-centric distances are perceived as compressed in immersive virtual environments relative to the real world. The principal factors responsible for this phenomenon have remained largely unknown. However, recent experiments suggest that when the virtual environment (VE) is an exact replica of a user's real physical surroundings, the person's distance perception improves. Based on this observation, it sounds reasonable that if subjects feel a high degree of situational awareness in a known VE, their ability for estimating distances may be much better compared to an unfamiliar virtual world. This raises the question, whether starting the virtual reality (VR) experience in such a virtual replica and gradually transiting to a different VE has potential to increase a person's sense of presence as well as distance perception skills in an unknown virtual world. In this case the virtual replica serves as transitional environment between reality and a virtual world. Although transitional environments are already applied in some VR demonstrations, until now it has not been verified whether such a gradual transition improves a user's VR experience. We have conducted two experiments to quantify to what extent a gradual transition to a virtual world via a transitional environment improves a person's level of presence and ability to estimate distances in the VE. We have found that the subjects' self-reported sense of presence shows significantly higher scores, and that the subjects' distance estimation skills in the VE improved significantly, when they entered the VE via a transitional environment.
Article
When an immersive virtual environment represents a space that is larger than the available space within which a user can travel by directly walking, it becomes necessary to consider alternative methods for traveling through that space. The traditional solution is to require the user to travel 'indirectly', using a device that changes his viewpoint in the environment without actually requiring him to move - for example, a joystick. However, other solutions involving variations on direct walking are also possible. In this paper, we present a new metaphor for natural, augmented direct locomotion through moderately large-scale immersive virtual environments (IVEs) presented via head mounted display systems, which we call seven league boots. The key characteristic of this method is that it involves determining a user's intended direction of travel and then augmenting only the component of his or her motion that is aligned with that direction. After reviewing previously proposed methods for enabling intuitive locomotion through large IVEs, we begin by describing the technical implementation details of our novel method, discussing the various alternative options that we explored and parameters that we varied in an attempt to attain optimal performance. We then present the results of a pilot observer experiment that we conducted in an attempt to obtain objective, qualitative insight into the relative strengths and weaknesses of our new method, in comparison to the three most commonly used alternative locomotion methods: flying, via use of a wand; normal walking, with a uniform gain applied to the output of the tracker; and normal walking without gain, but with the location and orientation of the larger virtual environment periodically adjusted relative to position of the participant in the real environment. In this study we found, among other things, that for travel down a long, straight virtual hallway, participants overwhelmingly preferred the seven league boots method to the other methods, overall
Conference Paper
The aim of Redirected Walking (RDW) is to redirect a person along their path of travel in a Virtual Environment (VE) in order to increase the virtual space that can be explored in a given tracked area. Among other techniques, the user is redirected on a curved real-world path while visually walking straight in the VE (curvature gain). In this paper, we describe two experiments we conducted to test and extend RDW techniques. In Experiment 1, we measured the effect of walking speed on the detection threshold for curvature of the walking path. In a head-mounted display (HMD) VE, we found a decreased sensitivity for curvature for the slowest walking speed. When participants walked at 0.75 m/s, their detection threshold was approximately 0.1m<sup>-1</sup> (radius of approximately 10m). In contrast, for faster walking speeds (>;1.0m/s), we found a significantly lower detection threshold of approximately 0.036m<sup>-1</sup> (radius of approximately 27m). In Experiment 2, we implemented many well known redirection techniques into one dynamic RDW application. We integrated a large virtual city model and investigated RDW for free exploration. Further, we implemented a dynamic RDW controller which made use of the results from Experiment 1 by dynamically adjusting the applied curvature gain depending on the actual walking velocity of the user. In addition, we investigated the possible role of avatars to slow the users down or make them rotate their heads while exploring. Both the dynamic curvature gain controller and the avatar controller were evaluated in Experiment 2. We measured the average distance that was walked before reaching the boundaries of the tracked area. The mean walked distance was significantly larger in the condition where the dynamic gain controller was applied. This distance increased from approximately 15m for static gains to approximately 22m for dynamic gains. This did not come at the cost of an increase in simulator sickness. Applying the avatar cont roller did reveal an effect on walking distance or simulator sickness.