Here is a draft of a short article I'm preparing for IEEE Pervasive
Computing magazine. I would appreciate suggestions on how to improve
it as it is intended for a technical audience who does not necessarily
know about wearable displays.
The image for Figure 1 is at
http://www.cc.gatech.edu/~thad/overlay-sv3.jpg
and the caption I'm suggesting is
Figure 1: Simulation of the overlay effect of using a monocular
opaque head-up display.
I was thinking of putting in a figure that helps illustrate the power
difference between PDAs and HUDs, but I think the analogy works well
enough and my drawing skills are poor. Opinions?
We could also put in an image of the SV-3 display being worn to go
with Figure 1 if people think it would add something.
Thad
------------
The Enigmatic Display
The monocular head-up display is the most distinctive component
of many wearable computers. Yet this style of display is often
misunderstood as to how and why it is used. This month I examine the
most common questions on wearable displays and point to new
research on their use.
*Why use a head-worn display?*
In many industries, workers have both hands occupied while requiring
access to information. For example, a surgeon may perform an
intricate microscopic procedure while watching a magnified view of his
actions on a head-up display. An overlay of the patient's vital signs may
provide the surgeon additional context during the procedure.
Similarly, a network technician might use a head-up display so that he
can monitor packet transmission while using his hands to physically
reconfigure a router. In Europe, BMW recently showed an augmented
reality (AR) system for automobile technicians; the head-up display was
used to overlay 3D graphics on a BMW engine which guide the technician
for each required step in the engine's service.
Beyond industry, monocular head-up displays also show potential in
consumer products. A growing number of individuals have adopted these
displays as part of their normal life. For me, the display provides a
quick way to access my calendar, refer to notes while teaching class,
inconspicuously write notes on a conversation, or even read the next
paragraph of an article while walking to my next appointment. I have
even been known to use a head-up display to read in bed so as to avoid
the fatigue of holding a book over my head for extended periods of
time.
*What does the display look like? What can you do with it?*
Figure 1 shows a simulation of using the MicroOptical SV-3 monocular
display. The SV-3 is a color VGA display with 640x480
resolution and a 16 degree horizontal field of view (19 degree
diagonal). In practice, the image from the computer seems to float in
space, overlaid on the real world. Because of a trick of the human
visual system, most users perceive that they "see through" the display
even though it is opaque.
A similar effect can be seen by holding your thumb a couple of inches
in front of one eye while focusing on something in the distance with
the other. The thumb is out of focus, of course, but you perceive
both your thumb and the object in the distance, even if the object
would at first seem to be obscured by your thumb. Closing first one
eye and then the other demonstrates how different the images to each
eye really are. Opaque head-up displays take advantage of this effect
to create the illusion of overlay. In addition, they use optics (much
like those of a microscope) so that both the display and the distant
object are in focus simultaneously.
Since many of these units can be driven from a standard VGA port, they
are capable of displaying information just like a normal desktop
computer. In general, monocular displays can be used by anyone with
normal vision, corrected or uncorrected. Some displays clip on to
eyeglasses or sunglasses if the user does not wear eyeglasses.
Other displays are mounted to a form of headband and sit forward far
enough from the forehead that the user can wear eyeglasses underneath.
--------------------------------------------------
TABLE/SIDEBAR
Advantages of head-worn displays:
Size/weight
Speed of access
Less vulnerability to damage
Hand support unnecessary
Less strain/fatigue for back, neck, and hands
Adjustable focus
Less power
Virtual overlay on physical world
Privacy
Less interruptive
Potential for large virtual image
--------------------------------------------------
*What are the advantages of a head-worn display?*
Compared to the displays of a laptop or PDA, modern head-worn
monocular displays have a distinct size and weight advantage. For
example, the head of my display weighs 35 grams (slightly over an
ounce). When I am not wearing my display, I store it in my shirt
pocket. In today's world, most people assume that it is an earphone
for a cellular phone or a MP3 player. However, when I need it, I can
still access the display very quickly.
This speed of access is another significant advantage. PDA users must
reach into their pocket or briefcase, uncase the PDA, boot it, pull
out the stylus, and get to the right application. With a head-up
display and a fast mounting system, the user can access their system
in as little as 1/10 of the time of a PDA user. Such accessibility
allows frequent use of the display and wearable computer for quick
reference and note-taking.
Since a head worn display is small and mounted near the face, it is
naturally more protected than a PDA or laptop screen. PDA screens are
relatively large and vulnerable surfaces in a mobile environment.
Since the PDA may be stored in a pocket, its screen can easily be
subjected to large forces if the user sits on it or places a large
object on his lap. Another problem with PDA-sized screens is that
they must be supported by a hand. In fact, most PDA interfaces
require the use of both hands (one for support and one for the stylus)
and both eyes. If a PDA user attempts to use the interface when
walking, much of his attention will be devoted to compensating for the
mechanical shock of the movement. Thus, both the user's physical and
attentional resources can be consumed by the PDA. On the other hand,
a wearable with a head-up display demands much less of the user
while mobile.
A head-worn display can also provide much better ergonomics than a
desktop, PDA, or laptop screen. Instead of requiring the user to sit
in an upright position with hands, neck, and back in the proper
location, the user of a head-worn display has much more freedom. I,
for example, often lay on the sofa in my office to write my papers.
Such freedom can be a release from occupational pain for sufferers of
back, neck, and hand injuries. (One unfortunate side effect, though,
is that visitors sometimes initially think I am sleeping when they
enter my office for their appointment).
Another ergonomic benefit of several of the modern head-worn displays
is their adjustable focus. Frequent users of desktops and laptops are
vulnerable to "computer vision syndrome" (CVS) due to the eye being forced
to attempt to focus at a near distance for extended periods of time,
Symptoms include headaches, loss of focus, burning/tired eyes, double
vision, blurred vision, and neck and shoulder pains. Some
optometrists even consider such computer use to increase the risk of
myopia (near-sightedness) in children. According to
Prio \cite{Priowebsite} and Bausch and Lomb \cite{Bauschwebsite},
makers of equipment for CVS, computer vision syndrome is due to the
difference in the resting point of accommodation (RPA) and the
distance users have to focus their eyes to read a computer screen.
The RPA is the distance at which the eyes focus by default (around 76
cm or 30 inches). With an adjustable focus screen, a head-worn display
user can vary his focus from a near depth to the RPA to even an
effective infinite depth depending on what provides the user with the most
comfort.
Head-worn displays also have an advantage over PDAs in the amount of
power they require. In order to be useable, a PDA must be viewable
from many different angles, even if the perceived image subtends the
same amount of effective visual arc as a head-up display. However,
a head-worn display is mounted so that its light is relatively
focused into the eye. Thus, head-worn displays naturally require less
power than a PDA screen. To illustrate the idea, the PDA might be
thought of as a flashlight which casts its light over a wide area
whereas the head-worn display might be thought of as a slide projector
that tries to provide a brilliant image in a limited area.
Head-worn displays offer some unique features over PDA and laptop
screens. Since the display is worn close to the eye, spying on the
user's screen without his knowledge is nearly impossible. Combined
with appropriate sensing, head-worn screens can be used to create a
real-time overlay of graphics on to the physical world. In addition,
accelerometers or a magnetic compass can be used to create a virtual
head-up display; as the user rotates his head, the image in the
display pans through a virtual image rendered in a ring around the
user's body \cite{billinghurst,macintyre}.
Head worn displays can also be less socially obtrusive than many
alternatives. For example, a cellular phone call can be announced
discretely with its caller ID in the user's display instead of an
insistent and uninformative mobile phone ring. As another example,
instead of needing to look away from his interviewee to put pen to
paper, a reporter can maintain eye contact while typing notes to his
display. Such subtlety can help avoid derailing the discussion as the
artifacts of the interviewer's notetaking (e.g. notepad, pencil,
writing, etc.) are no longer visible.
*If head-worn displays have so many advantages, why haven't they
penetrated the market?* or *Why aren't they here yet?*
While head-worn displays have a history going back to the earliest
efforts in computer graphics, only recently have small, mobile
computers and appropriate communications infrastructure been developed
that would benefit from their use. Now that the concepts of SMS
typing, game playing, and photography on cellular phones are becoming
popular, there will be greater impetus to adapt head-worn displays to
those markets. In addition, wide area wireless Internet access is
beginning to become reliable, and the public will begin to embrace the
idea that they do not have to be limited to their office to have
access to full-scale computing support.
However, microdisplay manufacturers still face difficult challenges.
For example, the display's field of view must be balanced against the
amount of visual area occluded by the display's support hardware.
Cost, brightness, contrast, power, resolution, social obtrusiveness,
and clarity are only some of the factors that manufacturers must
consider. While the state of the art has limited manufacturers in
these trade-offs in the past, recent studies have shown a beginning
maturity in the field, both in hardware and experimental practices.
In the Journal of Optometry and Vision Science, Sheedy and Bergstrom
report that users performed very similarly on paragraph reading,
letter counting, and word search tasks when using a monocular 800x600
resolution display (e-case by InViso) versus hard copy or a 15" flat
panel display. \cite{Sheedy} In some cases, the monocular display
actually outperformed the other display methods, but the results for
these tests were not statistically significant. Similarly, a
binocular display (e-shades by InViso) tested had slower or equal
performance rates on the tasks when compared to the monocular display,
hard copy, or the flat panel, but the results were not statistically
significant. In previous experiments, older head-worn displays did
not perform as well as desktop monitors or hard copy. In this study, the
authors attribute the current favorable comparison to improved display
resolution, partial instead of full immersion, and several other
effects best described in the original paper.
While Sheedy and Bergstrom's experiment shows the promise of the image
quality of the newer wearable displays, Laramee and Ware have been
exploring the effects of various backgrounds on task speeds when using
a monocular display. In their paper "Rivalry and Interference with a
Head Mounted Display," Laramee and Ware experiment with both a
see-through monocular display and an opaque monocular display. Users
were required to scan through a table of items and prices and use a
mouse to click on a specified price given a question such as "What is
the price of lettuce?" While performing this task, the users either
saw a bookshelf or a television playing a movie in the background.
The authors found statistically significant evidence for both
binocular rivalry (what one eye sees affects the other) and
interference (the background in a see-through display can conflict
with what the user is doing). However, the effects were not as strong
as the authors expected, especially in the case with the static
background. While there are many aspects that can be further explored
(adjusting brightness, contrast, and transpareny levels; using higher
resolution than the 450x266 IO Display Systems i-glasses in the
experiment; exploring focus effects with the TV background; exploring
other user tasks; etc.), this experiment shows a desire to examine
more complex tasks with head-worn displays. More such
experiments from the research community are needed to help display
manufacturers and wearable software providers tune their products and
overcome limitations to head-worn display use.
URLs:
http://www.prio.com/consumers/problem.shtml
http://www.bausch.com/us/vision/products/magnifiers/cvs.jsp
References:
James Sheedy and Neil Bergstrom. "Performance and Comfort on Near-Eye
Computer Displays." Optometry and Vision Science, 79(5), May 2002, pp.
306-312.
Robert Laramee and Colin Ware. "Rivalry and Interference with a Head
Mounted Display." ACM Trans. on Computer Human Interface (TOCHI),
9(3), September 2002, pp. 238-251.
Mark Billinghurst and Thad Starner. "Wearable Devices: New Ways to
Manage Information." IEEE Computer, 32(1), January 1999, pp. 57-64.
<need appropriate MacIntyre/Feiner article>
============================================================
--
Subscription/unsubscription/info requests: send e-mail with subject of
"subscribe", "unsubscribe", or "info" to 
Wear-Hard Mailing List Archive (searchable): http://wearables.blu.org
Please, *PLEASE* don't subscribe through a forward/expander/false domain
From Wear-Hard Mailing list Archive (WH)
Maintained by R. Paul McCarty
Archive created with babymail