Professional Documents
Culture Documents
Crossroadsxrds2015fall DL
Crossroadsxrds2015fall DL
ORG
APPRENEUR SCHOLARS
STUDENTS: SHOW US YOUR APPS
11
begin
5 LETTER FROM THE EDITORS
8 INBOX
9 INIT
Our Virtual Reality Revolution:
This time it’s for real
By Pedro Lopes
11 ADVICE
Writing your Dissertation
By Marius-Tudor Benea
12 UPDATES
Excellence Around the World
By Claudia Schulz
13 MILESTONES
Virtual Reality Device Developments
By Jay Patel
14 CAREERS
In Praise of Side Projects
By Natalie Gordon
16 BLOGS
From Novice to Expert:
The art of science of pedagogy
By Olivia Simpson
Top Image by A. and I. Kruk
Cover Art by
Iwona Usakiewicz / Andrij Borys Associates /
Shutterstock.com
18 46 64
features end
18 FEATURE 46 FEATURE 64 LABZ
L-to-R: Image courtesy of FOVE; Image by Durantelallera / shutterstock.com; Image of VFX-1 by eVRydayVR. Image of Oculus Rift Dev Kit 2 by Ats Kurvet.
One Step Beyond Managing Cybersickness The Virtual Human Interaction Lab
Virtual Reality: Connecting past in Virtual Reality By Andrea Stevenson Won
and future developments By Lisa Rebenitsch
By Diego Martinez Plasencia 65 BACK
52 FEATURE VR Head-Mounted Displays
24 FEATURE What Is the Vergence- By Finn Kuusisto
Substitutional Reality: Accommodation Conflict
Bringing virtual reality home and How Do We Fix It? 66 HELLO WORLD
By Adalberto L. Simeone By Robert Konrad Sorry Kids, Iron Man’s
and Eduardo Velloso Superpowers Aren’t Unique
56 FEATURE By Lara Zupan And Marinka Zitnik
30 FEATURE VR to the Rescue: Avoiding danger
Using Your Own Muscles: with mine disaster simulations 68 ACRONYMS
Realistic physical experiences in VR By Alain Boulay
By Pedro Lopes, Alexandra Ion, 68 POINTERS
and Robert Kovacs 61 PROFILE
Susumu Tachi: The scientist 69 EVENTS
36 FEATURE who invented telexistence
Eccescopy: To look, is to see By Adrian Scoică 72 BEMUSEMENT
By Ken Perlin
40 FEATURE
Lost in the Rift:
Engaging with mixed reality
By Daniel Boland and Mark McGill
LEARNING CENTER
Editor-in-Chief Mark Allman, Subscriptions ($19
Sean Follmer International Computer per year includes XRDS
Stanford University, USA Science Institute electronic subscription)
are available
Departments Chief Bernard Chazelle, by becoming an
Online Books
IPEC Ghaziabad, India Computing Machinery for profit or commercial
Marinka Zitnik 2 Penn Plaza, advantage and that copies
University of Ljubljana, Suite 701 bear this notice and the
New York, NY full citation on the first
from Safari, Books24x7, Slovenia
Digital Content Editor
10121-0701 USA
+1 212-869-7440
page or initial screen of
the document. Copyrights
Pedro Lopes C ON TA C T for components of this
Deep Dreams
are Made
of These
O
ver the summer, the Internet went into a collective frenzy for what looked like
psychedelic images more fitting for a late 1960s dorm room, not for the most
popular tech news sites. Images of dogs sprouting other dogs in fractal delight,
and other hallucinatory apparitions took over the Internet. I’m describing Google’s
“DeepDream” image software.1 It seemed like everyone and their uncle were posting crazy,
deepdreamed images on my Facebook feed for a few weeks. Of course these computers
weren’t dreaming, or at least not the way we do. A complex artificial neural network
trained on images, not REM sleep, are generated from unsupervised
is responsible for creating these learning, have existed for a long time
bizarre creations. (ANNs were hot in the 1980s), but now
UPCOMING ISSUES What began as a submission for the we have more data. In many cases the
2014 ImageNet Large-Scale Visual Rec- features the system finds make little
Winter 2015 ognition Challenge, became software or no sense at one level, but contrib-
created to identify objects and pat- ute meaningfully many levels up. But
[December issue]
terns. The DeepDream software takes now deep learning is starting to make
Internet of Things that code and optimizes input images big progress in speech and image rec-
so that its system can find features in ognition. Graphics and HCI research-
Spring 2016 images for easier classification. ers are starting to use these tools to
[March issue] Deep learning has been advancing produce exciting results; in a recent
the field of artificial intelligence at meeting, a colleague suggested using
Digital Fabrication
Google and beyond. These levels of ar- deep learning to classify some of my
tificial neural networks (ANN), which data. Once the HCI people start using
Summer 2016 it, you know it must be a big deal.
[June issue] 1 http://googleresearch.blogspot.
All of this comes out of an AI ap-
Cultures in Computing co.uk/2015/07/deepdream-code- proach that starts to look a lot more
example-for-visualizing.html like how our brain works than tradi-
COMPUTER SCIENCE
such freedoms?
Now is the time to think about
these issues. Can we hard code ethics
and morals into these robots? How
With a focus on network security, the MSCS positions students with can we teach ethics to robots if we
advanced technical skills essential to today’s programmers. can’t yet teach it to the engineers who
are programming the robots? With
• One-year or part-time formats • Competitive tuition large-scale engineering scandals
• Evening classes • Scholarships and financial aid like the Volkswagen Diesel exhaust
• Internship and networking • Optional test scores regulation fraud—the automaker was
opportunities caught with systems that falsified
pollutant numbers when tested—it
seems clear that when there are moti-
vations for financial gain ethical lines
become blurred. Before we can have
www.merrimack.edu/gradcs
ethical robots, we have to have more
graduate@merrimack.edu • 978-837-3563 NORTH ANDOVER, MA
2 http://futureoflife.org/AI/open_letter_autono- ethical engineers.
mous_weapons —Sean
— Follmer
Editor’s Note:
Thanks Natalie! Don’t
forget you can read the
latest issue online at
http://xrds.acm.org/
current-issue.cfm as well as
in the ACM Digital Library,
where you will find a link
for the digital edition. And
for those of you who like to
read on the go, download
our mobile app from the
iTunes store, Google Play,
and Amazon.
VOLUNTEERING WITH
XRDS Guessing bachelors/masters =portal&dl=ACM cfm?aid=2748105
@XRDS_ACM are you is the target? —Vaggelis Giannikas,
guys interested in HCI —Nicolai B. Hansen Research associate at the
/ Interaction Design University of Cambridge OTHER TWEETS
contributions :)? Editor’s Note: and Associate director of new in #XRDS_ACM
—Nicolai B. Hansen, Design If you are interested in the Cambridge Auto-ID Lab, The Wrens of Bletchley
researcher, Twitter (@ volunteering as a student Twitter (@VGiannikas) Park http://xrds.acm.org/
nbhansen) editor or blogger, please article.cfm?aid=2730914
contact us at xrds@acm. “I am also in the final stage. … #codebreakers
@nbhansen YES WE ARE. org with the subject line: Your article really helped #cryptography
What do you have in mind? “Student volunteers for me. Hope I have learned #engimamachine
—ACM XRDS Magazine, XRDS.” how to do good research #BletchleyPark
XRDS is ACM’s magazine for too. Thanks.” —Hacker Jou, Twitter
students by students, Twitter. Post by Ana Oliveira, Tue, (@hackerjou)
Image by Phish Photography
INIT
H
istory is repeat- are considering the world of
ing itself; we are VR, either as recent gradu-
reliving the verge ates ready to join the bub-
of virtual reality bling industry (think Valve,
(VR). The door is open and HTC, Samsung, Sony, Mi-
we can already peak in- crosoft, Facebook, Google)
side—through stereoscopic or as researchers looking
headsets made of plastic or to connect with fellow aca-
cardboard—into the magi- demics who research novel
cal realm of VR. Only time ways of experiencing VR.
will tell whether we will also Within these pages, XRDS
relive the death of VR. Al- shines light on both sides of
though computer scientists, the coin.
such as Ivan Sutherland, The University of Bris-
first crafted this technol- tol’s Diego Martinez
ogy in the 1960s, VR was Plasencia begins your voy-
later propelled by media age with a step-by-step nar-
hype in the 1990s. However ration of VR developments
it fell into a deep hiatus over the years. He guides
until 2012. It has only been readers from the designs of
three short years since VR consumer headsets, such
headsets experienced resur- as the Oculus Rift, to what
gence amongst consumers. might be the next, scene-
VR opens up an immense stealing technologies.
space where computer sci- VR technology allows
ence dives in; this special is- us to see new worlds, travel
sue will be an eye-opener for quickly to distant places,
emerging CS students who and be in full-body remote journey of “substitutional you to put on a very special
meetings. But as our eyes reality.” This is a technique type of virtual reality tech-
widen in surprise, the sen- that uses real-world objects nology: eccescopic goggles.
sation of “being there” is (think of all the everyday With this powerful tool,
broken as soon as we reach objects around you right imagined and provided
It seems out to touch anything. The
answer is immersive physi-
now) as stand-ins for vir-
tual objects. I also address
by Prof. Ken Perlin of New
York University, we embark
the last VR cal experiences, those that this challenge with my col- on a voyage that intercon-
revolution span beyond only “look-
ing at VR” and go as far as
leagues from the Hasso
Plattner Institute, Alex-
nects the shores of VR with
the continents of 3-D dis-
was missing “touching VR.” Working andra Ion and Robert Ko- plays, archipelagos of fu-
a critical on a solution to lost physi-
cality is Adalberto Sime-
vacs. We demonstrate new
mobile devices that enable
ture devices, and the poles
of human augmentation
component: one from the University of physical experiences in the philosophy and other ethi-
Photo Credit TK
In-depth.
unsolved VR challenges is to simplified solutions such
“cybersickness.” Lisa Re- as the Cardboard or the FO-
Innovative.
benitsch from Michigan V2Go, which leverage your
State University clarifies what existing smartphone as a
it is, what are the possible ex- VR display. Now we have
Insightful.
planations, and how can we makers, scientists, and the
expect to alleviate this con- public crafting VR applica-
dition for the future gen- tions; we are witnessing the
erations of VR technology. birth of new VR devices on
The VR Book: Human-Centered Robert Konrad of Stanford
University digs deeper into
Kickstarter every day, such
as input controllers (Six-
Design for Virtual Reality painful user experiences ense and Leap Motion) and
By Jason Jerald, PhD related to 3-D displays, and motion platforms (Virtuix
teaches us how the principle Omni and Vection VR). As
Good VR design requires strong communication between behind light-field displays of now, VR is starting to be
human and machine, indicating what interactions are might fix the “vergence- what we want it to be; since
possible, what is currently occurring, and what is about to accommodation conflict.” it is a constructed reality, VR
occur. A human-centered design principle, like lean At our final stop, Alain should be the place where
methods, is to avoid completely defining the problem at Boulay from Laurentian things are the way you
the start and to iterate upon repeated approximations and University shows us how want—your reality. Here, in-
modifications through rapid tests of ideas with real users. VR is used for altruistic ex- side the pages of XRDS, we
Thus, The VR Book is intended as a foundation for anyone share how that reality looks
and everyone involved in creating VR experiences today: from the challenges
including: designers, managers, programmers, artists, we face (cybersickness, lack
psychologists, engineers, students, educators, and user of haptics, etc.) to some
experience professionals. VR technology possible solutions (haptic
Available in hardcover, paperback and eBook. allows us to see wearables and light field
cameras). We at XRDS hope
DOI: 10.1145/2792790
new worlds, VR has a bright future as a
For more info please visit travel quickly tool for mankind’s augmen-
tation. The triptych formed
http://books.acm.org to distant by researchers, public, and
places, and be industry will undoubtedly
be instrumental to the suc-
M M in full-body cess of VR.
&C &C
Association for Morgan & Claypool
Computing Machinery Publishers
remote —Pedro
— Lopes,
meetings. Issue Editor
ADVICE
E
very student, no matter his or
her field of study, has a dream
for life post graduation. Nev-
ertheless, this dream is often
overshadowed by a legitimate fear of
the Ph.D. dissertation and its defense.
The best way to overcome this situa-
tion is to have confidence in the work
you present. If you want to experience
the certitude of well-done work, the fol-
lowing pieces of advice should be con-
sidered when working on a computer
science dissertation.
1. Start writing as soon as possible!
You will never have all the pieces you
need before beginning, but taking this
step early will create a fertile ground for
great ideas and for writing an excellent
dissertation. As soon as you have cho-
sen a subject and your supervisor(s), keep in mind there is someone will- hard to digest. To overcome this, use
don’t hesitate to get a head start before ing to help you. The person who is in many examples and figures and refer
you are too busy. the best position to do so is your su- to them whenever possible. Short code
2. Draft the manuscript. The best pervisor. Avoid, the common mistake examples, which are well integrated
starting point is to create a draft of made by computer science students into the narrative thread, can also be
the manuscript, including its desired everywhere: trying to do everything by used to help you make salient points.
structure, and filling in all the ideas yourself. Regularly ask your supervi- For larger blocks of relevant code, try
you have. This will give you and your sor for help and feedback. You will not to use appendices instead.
supervisor a good overview of the proj- only produce a stronger dissertation, 7. Acknowledge every contribution. It
ect and its progress later on. It will even you will also gain an advocate during is wise to note those who helped you,
spark your imagination for new ideas, its defense. and it would mean a lot to them. This
and will provide you with a “frame- 5. Review your manuscript regularly. will further improve your relationships
work” to easily integrate these ideas The previous advice doesn’t exempt in the long run.
into your dissertation while everything you from having to be honest with If you pay attention to all the tips
is still fresh. yourself. You should review the en- discussed, it is almost impossible to
3. Progressively integrate all your tire manuscript regularly. It will help fail. And the bonus is you will present
new ideas and results. Now, given the you develop confidence in the quality a well-regarded dissertation. Starting
above-mentioned draft of your manu- of your work. Once you discover new early, regularly integrating new con-
script, try to record every new idea or things and you have new ideas or re- cepts, checking your manuscript, and
result as soon as it comes to you. And sults, parts of the original manuscript interacting with your supervisor will
don’t forget to set aside time to do the will become outdated. Your work will not only assure your success, but it will
integration with the rest of the manu- need to be refreshed according to your also give you a better taste for the sub-
Image by A. and I. Kruk
script. After all, with these small steps new understanding of the studied ject studied. Be careful you might be-
and a small dose of perseverance you problem. come addicted. Don’t worry, this is one
can travel long distances. 6. Use many examples. Often the of those cases when such a situation is
4. Interact often with your supervisor. theoretical concepts introduced in desirable.
The best piece of advice is to always a computer science dissertation are —Marius-Tudor
— Benea
UPDATES
W
hat makes a student chap-
ter excellent? Is there a rec-
ipe for success? We turned
to this year’s ACM Student
Chapter Excellence Award winners
for answers.
University of Petroleum and Energy
Studies (UPES) ACM Student Chapter,
India. “We have a very disciplined and
specialized hierarchy,” explained Yash
Shanker Srivastava, joint head of the
events and workshop management
committee of the UPES ACM Student
Chapter. In his opinion, this is the
chapter’s secret for effectively orga-
nizing a large variety of outstanding
events. The activities that won them The winners of the Outstanding School Service award, the Tehran ACM Student
the Outstanding Chapter Activities Chapter, holding copies of F1 Journal.
award were mainly targeted at pro-
grammers—not just from UPES, but hour per day in the programming lan- cinnati ACM-W Student Chapter won
from all over the world. An example guage of the participant’s choice. This the Outstanding Community Service
is their “Hour of Code” event, which attracted more than 500 programmers award. The passion of Chawla and the
was held online and intended to be from five continents. The chapter’s chapter’s members is to spark an in-
an international programming chal- second outstanding activity had an terest in computer science. The chap-
lenge for coders (rather than an event even greater number of participants: ter collaborated with a local school,
for programming novices). The UPES more than 2,000 students across In- where they mentored and taught
student chapter set up seven program- dia took part in Prodigy 2015. The school children ways to tackle pro-
ming problems to be solved in seven chapter’s annual technical festival gramming projects. The chapter also
days; coding was scheduled for one featured a variety of coding competi- received sponsorship from Microsoft
tions, workshops on current IT trends, to organize DigiGirlz, an event attend-
and invited talks. “We never had an ed by more than 50 girls from Ohio.
aim for winning the award,” says Sriv- Their goal was to motivate more girls
astava, “we only aimed to do awesome to pursue careers in technology.
work for students and make them love University of Tehran ACM Student Chap-
The secret computing technologies.” ter, Iran. What made the University of
MILESTONES
Virtual
Reality Device
Developments
ment and typing on a Persian keyboard, Virtual-reality (VR) device designers have many
and programming classes for non-com- variables to consider when creating a new system:
puting students, to name a few. This use case, audience, comfort, and more. In this
meant members of the chapter could collection of milestones, we explore VR devices
“chose what they want[ed] to do and from different points in time and observe how their
whom they want[ed] to engage with,” creators chose to balance varying characteristics.
explained Amirreza Dadfarnia, editor-
in-chief of the chapter’s F1 Journal. This
quarterly publication contains not only
chapter news, but also articles covering
1950 Morton Heilig first describes his idea for
an “experience theatre,” which is an early
example of multimodal technology encompassing all five
hot topics in technology and computer senses into an immersive, cinema experience. Twelve years
science research. “Trust in yourself later, he builds the Sensorama prototype, which includes a
and never fear failure” is Dadfarnia’s display, odor producers, and a moving chair.
advice for becoming a successful
student chapter.
University of Karachi ACM Student
Chapter, Pakistan and Pontificia Univers-
1961 Philco Corporation develops the Headsight
device, a helmet that communicates with
a closed-circuit television system. It is intended as a
idad Javeriana ACM Student Chapter, Co- teleoperation device for the user to share findings when
lombia. The message is clear: In order exploring dangerous environments.
to develop excellence, a student chap-
ter needs a good balance of teamwork,
passion, and discipline. This is also the
advice shared by Madiha Nasim, the
1968 The Sword of Damocles is a head-mounted
virtual reality display, similar to goggles,
which uses a ceiling suspension system to support the
chair of the University of Karachi ACM heavy device on the user’s head. This system supports
Student Chapter, which won the Out- stereo images as well as field-of-view tracking.
standing Recruitment Program award.
“We work as a team, all with positive
attitudes, [...] for the betterment of the
society and after all computer science,”
1990 The first demonstration of Virtuality, a head-
mounted system for arcade-style games, is
released. Virtuality’s VR systems include sit-down pods and
explained Nasim. Alfredo Santamaría stand-up devices with joysticks for gaming.
agrees. He is the publicity leader of the
Pontificia Universidad Javeriana ACM
Student Chapter (for more informa-
tion on the chapter, see the Summer
2012 The first developer kit for the Oculus Rift, a
consumer-oriented head-mounted display,
is made available. Rift has received a great deal of industry
2015 issue of XRDS), which won the and consumer attention for its high-quality resolution, low
Outstanding Website award. His clos- latency, and affordability.
ing words of advice: “Give the best in
everything you do. ... The secret to our —Jay
— Patel
success is just not giving up under
any circumstances.”
—Claudia
— Schulz
CAREERS
G
etting your side project to
launch isn’t the end goal.
But not launching is the big-
gest reason that projects fail.
Learning to launch is one of the most
important skills you can develop to
better your career. Choose a project
that is as niche as possible. This way,
it’s more likely that you will actually
get to the hardest part (psychological-
ly), announcing “it’s live” and putting
it in front of real users. Yes you will be
vulnerable, and you will come up with
great reasons to push that uncomfort-
able feeling off, but launch when you’re
still embarrassed. Launch when you
know how incomplete your solution
is. You don’t need a “forgot password” shop. It doesn’t have to be code, but it I had previously launched a Spanish-
feature when you don’t have any users. does have to be something you create English language-learning site, which
You don’t need an awesome logo. Don’t from nothing. taught me first hand every lesson I
fall into the trap of thinking “you only I am four years into my latest side hope to teach you.
get to go live once, so it needs to be per- project. I’m the founder of an online Soon after my Hacker News post, I
fect the first time.” baby registry called BabyList. It start- met a young Filipino designer online.
Image by Eugenio Marongiu / Shutterstock.com
THE JOURNEY how your project will take the world by else, is the learning of the journey it-
Envisioning something and building storm. But once no one shows up, you self. I guarantee you’ll learn the les-
it yourself, with no boss or professor will be forced to hustle to attract us- sons I’ve mentioned. But first, launch
guiding you, is tough. But by narrow- ers by writing blog posts, answering it. Go public. Say, “I did this!”
ing your focus you can do something forum questions, working on SEO, and
meaningful for a small group of users. so on. Pro tip: Planning will never work
It’s easier to find a small group of us- as well as actually testing out ideas Biography
ers who really care about what you’ve post-launch. Natalie Gordon is the founder and CEO of BabyList,
built or are talking about it. Building ˲˲ You will fail many times. As painful an online baby registry. Before that, she graduated
from the University of Waterloo with a degree in computer
a next-generation social network is a as this sounds, it’s a necessary part of science and worked as a software developer at Amazon
terrible side project; although building real life. This isn’t a test you can study for five years.
BLOGS
focused on the acquisition of literacy skills: simple demonstrations. One striking research study conducted
reading, writing, and calculating. It was not the by Freeman et al. found overwhelmingly positive results
general rule for educational systems to train people among students who engage in active learning at least
to think and read critically, to express themselves
clearly and persuasively, to solve complex problems 1 Earlier this year, Bill Nye wrote an excellent commentary on the problem
in science and mathematics...More than ever, the for CNBC.com entitled “Fixing the US STEM Problem.”
25 percent of class time [2]. In fact, the results were so help with the more difficult concepts.
overwhelmingly positive there was immediate push to shift “A ‘metacognitive’ approach to instruction can help
research from whether or not active learning is effective to students learn to take control of their own learning by
which kinds of active learning are most effective. defining learning goals and monitoring their progress in
Peer instruction is one form of active learning that has achieving them.”
been adapted in a number of college classrooms. A basic Notice the two things mentioned here: defining
peer instruction cycle would work like this: learning goals and monitoring progress in achieving them.
1. Ask a multiple-choice question to the class. Good Defining learning goals is not necessarily a natural thing
questions are ones that prompt discussion and challenge for a student—especially a novice—to do. This is another
the preconceptions of students. way the educator can help. Explicitly defining and relaying
2. Each student will answer on their own. learning goals can be a great organizational tool for both the
3. Ask students to discuss their answers with their instructor and the student. It is important to have detailed
neighbors. It might be difficult to engage shy students, but learning goals at both a course and topic level, and to
often asking them to “convince your neighbor your answer explicitly address two or three with every class session. Good
is correct” get the conversation going. learning goals should involve verbs (and not something vague
4. After a few minutes, ask students to again answer on and unmeasurable like “understand”) and look something
their own. like “by the end of this lesson/unit/course, you will be able
5. Lead a class-wide discussion based on the correct to...” This provides students a way for self assessment; using
answers and the answers you saw. this framework of knowledge to monitor themselves in how
I have personally been in a class that used peer- well they are achieving their goals.
instruction in this way. The class was a subject enough Monitoring progress is slightly more complicated, and
outside my comfort zone that I actually needed that bit of should involve many levels of evaluation and feedback.
encouragement to engage in the discussion. I personally During class time it is important to give some feedback and
found having these little in-class quizzes encouraged me to to encourage a “try-fail-receive feedback-try again” cycle
prepare more for the class, and to openly talk through my and atmosphere for exploration. These small evaluations
thought process when answering a question. Ultimately, I felt should not be summative. It is always important for
I gained a much better handle on the material because of it. feedback to be timely and at an appropriate level. It is hard
“To develop competence in an area, students must: to achieve both things, but it equips students with the
a.) have a deep foundation of factual knowledge, appropriate sense of metacognition crucial for improving
b.) understand facts and ideas in the context of from novice to expert.
a conceptual framework, and c.) organize knowledge All of the above ideas are just a few of the new
in ways that facilitate retrieval and application.” techniques being tested and used by instructors in
I think the most important point in this key finding is classrooms all over the country. These techniques could
point c—organizing knowledge in a way that facilitates really transform a classroom for the better—especially
retrieval and application. This is the point at which in STEM classrooms that are not especially discussion-
students step beyond learning facts and move toward based. The STEM problem is not insurmountable, and
learning tools and strategies. One way educators can instilling excitement and curiosity in students at an early
facilitate is by teaching some matter in depth, but age will encourage many to go on to get degrees and
providing many examples of the same concept at work. develop the skills this country needs.
Of course, there must be a firm foundational knowledge
base upon which the students organize information. A References
big question I’ve encountered among educators is how [1] National Research Council. How People Learn: Brain, Mind, Experience, and School:
to use a finite amount of class time to both impart the Expanded Edition. Washington, DC: The National Academies Press, 2000; http://www.
nap.edu/openbook.php?record_id=9853&page=1
factual knowledge base and build facts into a conceptual
[2] Freeman et al. Active Learning Increases Student Performance in Science,
framework. One idea is the “flipped classroom” model. Engineering, and Mathematics http://www.pnas.org/content/111/23/8410.abstract
In this setting, students are asked to do reading or small
homework exercises before class time. The idea is to ask Biography
students to learn the easy things on their own at home Olivia Simpson is a Ph.D. candidate at the University of California, San Diego working on
(definitions or basic concepts), while using class time to algorithms for network analysis.
V
irtual reality (VR) has gained incredible public visibility in the last three years,
undoubtedly as a result of Palmer Luckey’s vision: the Oculus Rift. What initially
started as a hobby, soon took off, becoming one of the brightest examples of
crowdfunding success. Over the years, Luckey accumulated a large collection of VR
headsets and by the age of 16 he started building his own. On August 1, 2012, he launched
his Kickstarter campaign, pledging a modest $250,000 and aiming to provide cheap and
high-quality headsets to a few VR enthusiasts. Unexpectedly, the campaign raised more
than $670,000 in its first 24 hours, and more than a million dollars within three days.
But Luckey’s most amazing achievement was making the world excited about VR again.
After witnessing the rise and fall The cinematographic industry also fu- develop a VR headset with unforeseen
of so many companies intent on com- eled our expectations to unreachable quality and priced under $300.
mercializing VR in the late ‘80s and levels with their futuristic vision of 3-D From that moment on Oculus has
‘90s (older readers might remember interfaces. After this long and continu- been a runaway success, which ended
the excitement around the Virtuality ous trickle of disappointments, VR en- with Facebook acquiring the company
gaming system, or how you felt when thusiasts were left heart-broken, hope- for $2 billion last year. The rise of Ocu-
you played “Dactyl Nightmare” for the less, and extremely skeptical. lus has also dramatically reshaped the
first time in an arcade). After decades However by the end of 2012, Ocu- stage of the technological market. Re-
of expensive and disappointing head- lus released its first development kit cent market reports foresee VR becom-
mounted displays (HMDs), I still re- only a few months after its campaign ing a mainstream technology in the
member my first experience with a Kai- launch—a big statement against all of near future. Last November, analysts at
ser ProView XL35. It was priced at more this skepticism. Although it was not Sophic Capital predicted VR will reach a
than €20,000 and only rewarded its yet a consumer-ready product, Luckey $7 billion market by 2018. Others, such
user with a field of view of 35 degrees. clearly demonstrated it was possible to as Piper Jaffray, estimate up to $62 bil-
proposed cases where users can fit their esting features to the plethora of com- display quality and gaze tracking ac-
smartphones to create VR experiences mercial VR headsets. In particular, curacy, maybe at the expense of com-
on the go. Valve’s bet is called HTC Vive FOVE includes gaze tracking in their fort. It felt slightly heavier than Ocu-
and aims to bring VR to its huge com- headsets to determine where you are lus’s DK2. Another ambitious project
munity of gamers. Sony’s Project Mor- looking, allowing possibilities out of is Mind Maze VR, which integrates an
Images from E3 2015 courtesy of Entertainment Software Association (ESA). Images from CES 2015 courtesy of Credit Consumer Electronics Association (CEA).Razer OSVR image courtesy of Maurizio Pesce (Flickr).
throughout the day. Whether these
potential developments outweigh the
impact (on your peripheral vision and
social life) of having the Hololens cover
your face, is yet to be seen.
REVOLUTION OR REFURBISHMENT?
(a) (b) The resurgence of AR/VR hardware
clearly demonstrates technology has
reached a point, in terms of quality and
price, where these devices can be made
available to the public in an attractive
format. Mobile devices can satisfy the
computing and power necessities of
these resource-hungry devices, while
remaining cheap and lightweight. Dis-
(c) (d) play size, refresh rate, and resolution as
well as the advances of sensing technol-
ogies at affordable rates (such as Kinect,
array of electroencephalography (EEG) The situation for newer AR headsets Leap Motion, GloveOne, and a myriad
sensors and also features a depth cam- was different. With all of the hype sur- of other devices) have reached a level of
era on the front of the device. The cam- rounding Oculus-like headsets, later maturity that makes the current offers
era supports hand interactions and AR solutions were conceived free of the possible. The spark required to actually
self-localization of the head. Interest- aesthetic concerns that constrained push technological giants to take on
ingly, it allows for augmented/mixed Google Glass. These AR headsets em- this challenge for consumer grade VR
reality experiences, which are experi- brace solutions that cover the user’s grew stronger with Luckey’s Kickstarter
ences that mix the virtual environment face, but allow for AR to be explored in campaign. After years of seeing VR/AR
and real world located around the user. full depth. Some of the earliest exam- through the big screen, people want to
Another consequence of this small ples are OVRVision and Leap Motion, experience it in their homes
revolution: Oculus’ momentum man- which both proposed clip-on solutions However, although the current
aged to also bring augmented reality to add a stereo video feed of the real range of devices undoubtedly outper-
(AR) back to life and into the market. world for Oculus Rift. VRVana’s Totem form the devices available before, it is
This adoption was initially slower, and Impression Pi were probably the also true that the formats proposed are
maybe influenced by what had been— first to add built-in cameras to their in no way different than the ones ex-
at best—a slow uptake of Google’s at- headsets, which provided integrated plored during the 1970s and 1980s by
tempt in this field. Although Google support to AR. the research community. Back in 1968,
Glass was conceived before Oculus ar- The big leap arrived at E3 2015, when Ivan Sutherland—who was a pioneer
rived, at the time skepticism against Microsoft presented its prototype of a in computer graphics, AR, and VR—
intrusive devices was strong in the pub- see-through AR headset: the Hololens. built the first head-mounted display
lic’s mind. Google attempted to keep Unlike the previously mentioned de- that included stereo 3-D graphics and
their device unobtrusive and aestheti- vices, Hololens features a semitrans- head tracking. In the following years
cally pleasing. But its small display, lo- parent screen that allows you to see the different types of headsets, tracking
cated off to the side, did not allow wear- real world with your own eyes, instead technologies, tactile and force feed-
ers to explore the full potential of the of a digitized video feed, which, even back devices, data-gloves, and 3-D con-
AR experience. This coupled with its if stereoscopic, would be no match for trollers were tested. According to this
not-so-attractive price (approximately the accuracy of human vision. Virtual older research, recent devices are not
$1500) and some social concerns (such contents can then be added onto your new at all. Today’s technology can eas-
as surreptitious video recording), pos- view of the real world, either in subtle ily be categorized according to histori-
sibly led Google to abandon the project ways—to help you find your way as you cal terminologies: Oculus, Gear VR, or
earlier this year. walk through a city—or overriding the HTC VIVE (and their 3-D controllers)
Figure 3a. BaseLase from Aarhus University. Projectors with the ability to cover big areas and even to be integrated in
public furniture are becoming available. [Reprinted with permission from Aarhus University.] Figure 3b. RoomAlive from
Microsoft Research. Interactive walls and projection are one of the alternatives to VR that allow ubiquitous augmentation
of the object and spaces we inhabit. Figure 3c. IlumiRoom from Microsoft Research. When coupled with head tracking,
projectors can be used to create immersive 3-D effects, without isolating the user from the real world. [Reprinted with
permission from Brett Jones].
Figure 3a). Mobile devices with embed- the correct perspective as the user to be used as windows to the AR ex-
ded projectors have been proposed as moves can create compelling 3-D illu- perience. The information projected
an alternative to display information sions. The Microsoft Research team from each user is retro-reflected to his
on the objects around us. In this case, lead by Andy Wilson has demonstrated or her eyes, so it only becomes visible
the device is bound to the user instead a variety of projector-based systems ex- to the user. CastAR also provides the
of a fixed infrastructure, but is still ploring this concept. In particular, the possibility to correctly see and inter-
part of the IoT. Beamatron combines a depth camera act with the 3-D contents using your
All these approaches can overlay and a projector on a pan-tilt platform. hands or controllers, which is not
augmentations (notifications and infor- This allows the system to track the possible with traditional optical see-
mation) on the world around us and not user’s head, the room, and project—for through HMDs. These headsets pres-
occlude users’ faces. However, there are instance, a racing car that interacts ent content floating at approximately
two important limitations when com- with the objects in the room. Illumi- 3 meters in front of us. Our eyes (we
pared to AR/VR headsets: 1) augmenta- Room and RoomAlive demonstrate can think of them as a lens with a usu-
tions lay on the surface of the world, you this concept with 3-D augmentations al depth of field of + 0.3 diopters) can
cannot create a virtual pet that seems popping out from any point on a wall or blend content with objects, which are
to walk on a table; and 2) information is even a whole room (see Figures 3a and more than 1.6 meters away from us.
equally visible to everyone. 3b). Taking a more literal approach, Thus, we will be able to blend the aug-
Overcoming the first limitation— the Tangible Media Group at MIT’s mentations on road signs and cars as
creating illusions that pop out from Media Lab created a shape-changing we drive (hopefully more than 1.6 me-
the surface where they are projected— tabletop, whose shape changes to ap- ters in front of us). However, if our eyes
usually requires stereo techniques. proximate that of the virtual content to focus on our finger 40 centimeters in
However, stereo is an effect that decays display. front of us, the contents of the head-
greatly as objects become more dis- CastAR, from Technical Illusions, set (optically, at 3 meters) will appear
tant. Thus, detecting the user’s head combines projected AR with retro- blurry. In contrast, CastAR augmenta-
position and showing the object from reflective materials on the surfaces tions’ depth is that of the table or wall
Figure 4. Systems embedding displays made of fog provide a look and feel similar to the headset based VR. Figures 4a and 4b.
Mistable allows users to interact with their personal contents on a fog screen in front of them. These contents can be shared on
the central tabletop, or they can be pulled above the tabletop to reveal their 3-D appearance and interact with them. Figure 4c.
Sensabubble uses projectors and soap bubbles filled with smoke to create displays that can float anywhere around the room.
PiVOT is the first system that com- air. This arrangement allows users ac-
bines Lumisty film, a projector, and cess to the shared tabletop and to see © 2015 ACM 1528-4972/15/09 $15.00
Substitutional Reality:
Bringing virtual
reality home
Now that virtual reality headsets are finally reaching the wider
consumer market, how can we merge the physical and virtual worlds
to create a unified multi-sensory experience?
F
or many of us, virtual reality (VR) evokes all sorts of imagery—from neon vector
graphics to dream worlds built to imprison humankind. In its essence, VR as an
artistic medium promises to fulfill a quintessential human desire of exploring places
and living lives that are unattainable in the real world. VR has also found fruitful
practical applications in the fields of architecture, training, design, and rehabilitation. And
with the release of consumer head-mounted displays, such as the Oculus Rift and other
headsets, VR is finally ready to enter our homes. The textbook definition of VR refers to the
artificial simulation of sensory experiences. Modern VR systems do a great job of simulating
visual and auditory sensations, but there is still a long way to go until we are
able to successfully simulate the other real-world counterpart that the user are significantly different than our
senses, often described as the “holy could experience physically? We call furnished (and often messy) homes.
grail” of VR. To truly understand the this “substitutional reality” (SR) [1]. For example, the instruction manual
potential of VR, it is necessary to expe- Each virtual object has a physical ob- for HTC’s Vive headset explicitly rec-
rience the complete package. We are ject that substitutes it, allowing users ommends users to rearrange furni-
interested in looking for alternative to experience the virtual world in the ture in order to make space for the
possibilities for stimulation without same way as the physical world. They tracked area. Other home VR systems
necessarily simulating experiences can navigate the environment without often avoid this problem by limiting
through additional actuators. Our bumping into furniture, or touch and users to their desks, and allowing
approach is to leverage the physical feel virtual objects as if they were real navigation through mice, keyboards,
properties of the world around us to (see Figure 1). and game controllers. However, this
Image by Jennifer Morrow
stimulate the senses that we cannot Research labs responsible for effectively removes two of the most
yet simulate in VR. What if physical building VR systems often employ compelling presence-enhancing fac-
objects, rather than being obstacles large, empty rooms that allow users to tors in our living rooms—touch and
or impediments, were incorporated move freely without the fear of bump- locomotion.
into the virtual environment? What if ing into furniture or tripping on ob- Previous work on passive haptics
every object in the virtual world had a stacles. However, these environments has shown the use of an exact match
Figure 1. Two virtual environments based on the layout of a real room (middle), a medieval courtyard (left), and the bridge
of a spaceship (right). In these substitutional environments, every physical object is paired, with some degree of discrepancy,
to a contextually appropriate virtual object.
between the physical and the virtual which it is deployed. There are three face reconstruction, object tracking,
objects can highly benefit the experi- main components to an SR system. and 3-D rendering at a frame rate high
ence [2]. However, if virtual reality is a First, it requires a means of captur- enough not to cause nausea is a highly
mere replica of the physical world, is ing the physical environment. The computationally intensive task. The
it easier to just take the headset off? most basic way of accomplishing this feasibility of these components has
In SR we seek to embrace this mis- is to manually create a virtual envi- been previously demonstrated in
match between the two environments ronment that substitutes the physical other works and products [3]. A Utah-
and investigate how far a designer can one. Individual objects can then be based startup, The Void, is working
push this difference until it breaks tracked using computer vision, infra- on an SR experience using a theme
the illusion. What if a designer wants red markers, or inertial measurement park approach. With a combination
to add objects that do not have a physi- units. More sophisticated approaches of props, wearable haptic devices, and
cal counterpart? include scanning the environment environmental effects, the company
Imagine you are in your living with a wearable depth camera or even proposes to create multisensory expe-
room, sitting on your couch, and combining the input from several riences. In “the Void,” a physical envi-
drinking a cup of tea while watching cameras using projection mapping. ronment is specifically built to match
TV. You turn off the television, place Second, it requires a means to rec- the virtual environment. The physical
your mug on the coffee table in front ognize, segment, and map physical props provide haptic feedback and
of you, and put on a VR headset that objects to their virtual counterparts. the system uses tracking data to pair
transports you to a medieval court- Possibilities range from supporting them to their virtual counterparts.
yard. All of a sudden, the couch under- users to manually creating such map- In substitutional reality, users are
neath you becomes a wooden bench. pings to automating the substitution fully immersed in the virtual environ-
The tea cup is now a ceramic mug rest- process. Third, it needs to render the ment. However, other works in mixed
ing on a beer keg. Where there used virtual environment back to the user reality also aim to merge the physi-
to be a TV, you now see a sword rack. through a head-mounted display cal and virtual worlds. For example,
Every object in the physical world is while it tracks moving objects. This Microsoft’s Hololens overlays digital
now substituted by a new one, which can present significant challenges for content onto physical objects through
matches the context and theme of the system developers as combining sur- a wearable see-through display. On
virtual environment, but with certain the other hand, RoomAlive, also by
variations. Would you still believe you Microsoft, projects digital content
are sitting on a wooden bench despite onto the physical environment using a
the comfort of a cushioned couch? Every object in the combination of projectors and depth
Would the difference in size between
the teacup and the beer mug affect
physical world is cameras [4]. Although these projects
have different motivations and goals,
how you manipulate it? Would you now substituted by the underlying techniques can also be
still believe you were drinking beer
as the sweet taste of tea touches your
a new one, which used for SR systems.
lips? These are only some of the ques- matches the context LEVELS OF MISMATCH
tions that this scenario uncovers.
and theme of the Considering the mismatch between
physical objects and their virtual
ENABLING TECHNOLOGIES virtual environment, counterparts, we propose a sequence
An SR system is capable of dynami-
cally creating a virtual environment
but with certain of five levels of mismatch. At the bot-
tom level we have replicas, substitut-
based on the physical environment in variations. ed objects that are 1:1 representations
Figure 2. In the picture each pair of objects exemplifies a different substitution type. The replica mug is used as baseline
(a). The pair of substitutive objects are: (aesthetic) glass (b) and wooden mug (c), and hot (d) and icy mugs (e); (addition/
subtraction) big mug (f) and small mug (g); (function) basket (h) and lamp (i); and (category) box (j) and sphere (k).
ate representation can then be adapt- sive feedback, with the potential of [1] Simeone, A. Substitutional Reality: A research
agenda. In Proceedings of the first Workshop on
ed to the particular requirements of heightening the sense of presence. Everyday Virtual Reality (WEVR) at VR 2015. IEEE,
the VR experience: The bounding box- In automated mapping of virtual to 19-22.
es identifying a chair can be turned physical objects, designers of SR expe- [2] Hoffman, H. G. Physically touching virtual objects
using tactile augmentation enhances the realism of
into a car seat or a tree trunk, a desk riences will have to do so without ac- virtual environments. In Proceedings of the Virtual
Reality Annual International Symposium, IEEE,
into a stone altar, etc. This is no easy tually knowing what the physical en- 1998,59-63.
task and will require research into vironment looks like. The logic of the [3] Izadi, S., Kim, D., Hilliges, O., Molyneaux, D.,
both hardware and software solu- experience needs to be described in Newcombe, R., Kohli, P., Shotton, J., Hodges,
S., Freeman, D., Davison, A., and Fitzgibbon, A.
tions. We believe it can be achieved in abstract terms, i.e., “place this object KinectFusion: Real-time 3D reconstruction and
steps: Starting from the dynamic sub- near a window if available, otherwise interaction using a moving depth camera. In
Proceedings of UIST 2011. ACM, 559-568.
stitution of the immediate surround- on the first available planar surface.”
[4] Jones, B., Sodhi, R., Murdock, M., Mehra, R., Benko,
ings of a user’s desk, then proceeding Further, it is conceivable that for the H., Wilson, A., Ofek, E., MacIntyre, B., Raghuvanshi,
to the substitution of a single room, same experience not all physical en- N., and Shapira, L. Roomalive: Magical experiences
enabled by scalable, adaptive projector-camera
and eventually large environments— vironments will present the same de- units. In Proceedings of UIST 2014. ACM, 637-644.
perhaps even outdoor ones. The first gree of mismatch. For example, if we [5] Simeone, A., Velloso, E. and Gellersen, H.
step especially supports what the consider a medieval experience, the Substitutional reality: Using the physical
environment to design virtual reality experiences.
gaming press regards as a “killer ap- degree of mismatch will be a lot less In Proceedings of CHI 2015. ACM, 3307-3316.fdfd
plication,” specifically cockpit games if deployed in an actual castle rather
such as flight and racing simulators. than in a high-rise apartment. How Biographies
These VR experiences simulate simi- can we quantify the mismatch in or- Adalberto L. Simeone is a lecturer at the School of Creative
Technologies at the University of Portsmouth (UK). His
lar conditions to those we would find der to objectively compare the same research focuses on how to bridge the gap between users
in reality. In most cases, it would not experience across different physical and their access to 3-D interaction and virtual reality
technologies.
be possible to walk around the cock- environments?
Eduardo Velloso is a Ph.D. candidate at Lancaster
pit and thus these are ideal for a desk- University and will take up a position as a Research Fellow
top VR game. One of the disadvan- CONCLUSION at the Microsoft Research Centre for Social Natural User
Interfaces at the University of Melbourne in January 2016.
tages of VR is we lose awareness of the These are exciting times for virtual His research interests include multi-modal interaction,
surrounding environment while im- reality. Although it failed to become eye tracking, and post-desktop user interfaces.
Using Your
Own Muscles:
Realistic physical
experiences in VR
Leveraging the user’s own muscles to simulate impact and forces
from a virtual reality world allows us to create more immersive
experiences without bulky equipment.
C
onsumer virtual reality headsets (such as the Oculus Rift, Samsung Gear VR, or
Google’s Cardboard) provide an exciting peek through the window of virtual reality
(VR), but that is where our experience ends. As we reach out to touch these novel
realities we are left with a sense of disillusion, because all we see are 3-D ghosts
without any mass. To give VR users a sense of an object with mass, traditionally large robots
are used that can, for instance, move the user or simulate virtual terrain. However, these
large and stationary robots are out of place in the mobile age: We are used to small and
light, yet powerful, devices in our pockets. After all, consumer-ready VR became a reality Image by Alicia Kubista / Andrij Borys Associates / Shutterstock.com
when the headsets finally met the size and portability requirements of our mobile times.
At the Hasso Plattner Institute’s their fingers, arms, and even the ter- as Lufthansa’s flight simulator.1 These
HCI Lab, our research has focused rain under their feet. This is our con- motion platforms are heavy station-
primarily on bringing a physical tribution to making virtual reality ary machinery that are too impractical
component to mobile VR experi- physical. to ever reach our homes—even if you
ences. Users can reach out and feel could afford a large motion platform,
what is, virtually, there. Because we MOBILE TERRAIN SIMULATION space is sacred these days. This tech-
have engineered these devices to be From the inception of VR in the 1960s nology was acceptable when VR was
wearable, they leave the user uncon- [1], through the industry hype of the a specialist product, and these ma-
strained to freely move in space and 1980s, up until today, many efforts chines shone through their precision
enjoy the virtual experience to its have been geared toward helping users rather than their form factor. In fact,
fullest. In this article we describe a to “feel” virtual worlds. Industry and the biggest players in VR were not gam-
new world of VR, in which users feel research labs responded to this chal-
forces from the virtual world through lenge by using motion platforms, such 1 http://www.virtual-fly.com
THE CHALLENGE OF
MOBILE FORCE FEEDBACK
Level-Ups was, for us, a clear step in the
right direction toward mobile physical
VR, but is it what sci-fi had us dream-
ing about for years? What about feeling
forces from the virtual world?
These dreams of a physically be-
lievable virtual reality aren’t new, but
so far they required heavy gear, such
ers, but training centers (e.g., airplane piece of infrastructure such as Circu- as exoskeletons. These are large me-
pilots) and amusement parks (e.g., laFloor. It is enough having a Roomba chanical contraptions mounted onto
VR rides). roam around a house for cleaning pur- users; a motor is placed at every joint
These massive motion platforms poses, but four jumbo-sized Roombas of their limbs and thus users are able
were able to provide the sense of “be- moving around the living room for the to manipulate their limbs while they
ing moved” or the sensation of accel- sole purpose of emulating VR terrain? experience VR. Think of motion plat-
eration, but weren’t really suited for In one of our projects, we took a dif- forms that you carry around. The hu-
simulating a “walk in the park.” Thus, ferent approach to ground simulation. man is rather powerful, and therefore
many researchers designed mechani- Instead of placing the human into a the motors required to move our limbs
cal devices targeted at rendering the robotic device, we mounted motors must be powerful, too. And powerful
sensation of stepping onto uneven under the human. With this approach, motors are large. Now if you look at
ground. Canonical examples of these the physical feedback device goes Google Cardboard, Oculus Rift, or any
were treadmills equipped with indi- wherever the human goes—it is mo- other headset on the market, you real-
vidually height-adjustable elements bile. Our computer-controlled stilts, or ize their form factor is orders of mag-
that simulated bumpy terrain and “Level-Ups,” allow users to physically nitude smaller than the actuator sys-
virtual slopes [2], or the CirculaFloor, experience elevation. Level-Ups are tems that would allow us to bring some
which uses four robot units that place designed to allow users to freely walk physicality to virtual experiences. So
themselves under the user’s steps [3]. in the virtual and physical world, and how can we produce enough power to
However, in the era of consumer-ready produce a higher sense of presence. actually move a human limb, but in a
mobile VR we cannot expect to own a While you walk in a virtual world very concise form factor?
2 http://www.teslasuit.com/
References
[1] Iwata, H., Yano, H., and Fukushima, H. CirculaFloor:
A locomotion interface using circulation of movable
tiles. Computer Graphics and Applications, IEEE 25, 1
(2005), 223–230.
[2] Noma, H., Sugihara, T., and Miyasato, T. Development
of ground surface simulator for Tel-E-Merge system.
In Proc. VR ‘00. IEEE, 2000, 217–224.
now image that all those impacts be- ence: soccer juggling. This setup points [3] Strojnik, P., Kralj, A., and Ursic, I. Programmed
come real, and you feel them in your the solenoid component at the user’s in- six- channel electrical stimulator for complex
stimulation of leg muscles during walking. IEEE
skin and muscles. step (top of the foot) and the EMS unit to Trans. Biomedical Engineering BME-26, 2 (1979),
To demonstrate how one could feel the calf muscles. The electrical muscle 112–116.
impact in VR experiences, we imple- stimulation causes the foot to slightly [4] Sutherland, I. E. A head-mounted three dimensional
display. In Proc. AFIPS ‘68. ACM, New York, 1968,
mented three VR sport simulators: bend backward at the moment the ball 757–764.
boxing, soccer juggling, and baseball. hits the foot. Also, by placing impacto [5] Tamaki, E., Miyaki, T., and Rekimoto, J. Possessed
hand: Techniques for controlling human hands using
Boxing is a sport in which impact units on arms and legs, users experience electrical muscles stimuli. In Proc. CHI ‘11. ACM, New
sensation is crucial. The opponent’s av- a Thai boxing simulator where they feel York, 2011, 543–552.
atar keeps its guard up and attacks pe- virtual punches and kicks. As in all our
riodically. Users must choose the right sport simulators, we use a Kinect cam- Biographies
moment to start a successful attack. era to track the user’s body posture and Pedro Lopes is a Ph.D. student of Prof. Patrick Baudisch’s
Human Computer Interaction lab in Hasso Plattner
When the virtual character attacks, the an additional wireless accelerometer on Institut. Lopes creates wearable interfaces that read
impact of the virtual punch is rendered the solenoid to determine the foot’s tilt. and write directly to the user’s body through our muscles
[aka proprioceptive interaction]. He augments humans
with impacto—users feel the tactile Lastly, we can actuate users, even and their realities by using electrical muscle stimulation
sensation on the location where the while they operate objects. This allows to actuate human muscles as interfaces to new virtual
worlds. He also enjoys playing improvised music and is the
avatar hits them and they feel their arm impact sensation to happen in hand- digital content editor of XRDS.
being thrust backwards due to the mass held props that the user manipulates Alexandra Ion is a Ph.D. student of Patrick Baudisch at the
of the avatar’s arm colliding with their in a virtual world: Imagine hitting a Human-Computer Interaction lab at the Hasso Plattner
Institute. She is interested in haptics and mechanical
own arm. When users attack, again we baseball with your VR bat. Figure 5 properties in fabrication. Ion has previously worked with
render the impact by tapping users at shows a prop becoming animated; Dr. Michael Haller at the University of Applied Sciences
Upper Austria, Campus Hagenberg on large interactive
the location they encountered the oppo- here a stick is a stand-in for a baseball whiteboards with pen input.
nent, e.g. their fist or arm, and stimulate bat. We mounted the solenoid onto Robert Kovacs is a mechatronics engineer and artist.
their biceps causing their arm to stop the prop but the EMS unit, in contrast, Currently, he is a Ph.D. student at the Human-Computer
Interaction lab at Hasso Plattner Institut and works
upon collision. stayed with the user; here it stimulated on large-scale fabrication. He previously studied at
Users can use multiple impacto units the wrist muscles exactly when the ball Free University in Berlin and the University of Novi Sad,
Serbia. Kovacs is the co-founder of Oktopod Studio,
at the same time, as well as mounting hits the virtual bat. an educational tool for mechatronics, robotics, and
them on other limbs and muscles (such automation based on open-source hardware.
Eccescopy:
To look,
is to see
Creating a future where dreams walk among us.
By Ken Perlin
DOI: 10.1145/2810052
I
n 2003, I visited Will Wright at Maxis, while he was still working on “The SIMS 2.”
He showed me a box—exactly the size of a computer game CD box—with nice artwork,
text, and system requirements. It was everything you’d expect, except it was labeled “SIM
Everything.” The release date was 10 years in the future. I looked more closely at the system
requirements, and they were far beyond anything that was available then. Will explained this was
always the way he and his colleagues planned new game releases. Right up front they design the
box, the artwork, the characters, and the nice little blurb that goes on the back of the box.
Turning the CD box over in my hand, I world. Each of us will have our own ECCESCOPY IN POPULAR CULTURE
said, “So the box is actually empty?” “No,” personal view of this augmented re- Of course, the dream of merging the
he replied, “The game is already in the ality. I call this sort of display “ec- real and the virtual is far from new. Vi-
box. You just can’t open it yet.” cescopic,” from the Latin “ecce” and sions of an eccescopic reality have a
A few years later I read Vernor “scope.” To look is to see. long history in popular culture. These
Vinge’s novel, Rainbows End. I re- Will it be a good thing or a bad visions often highlight real killer apps
alized everything in it would be at- thing when virtual objects appear for eccescopy: Enhancing the ability of
tainable in the next few decades. to inhabit physical space? And how people to communicate with each other
Computer screens will become su- will it affect our relationship with without disrupting their sense of shared
perseded by wearables and eventu- the world around us? Will eccescopy physical space.
ally by contact lenses. People will be- take us even further away from physi- For example, in the 1957 film “For-
come used to seeing virtual objects cal reality, or will it allow us to better bidden Planet,” a machine developed
superimposed onto the physical join mind and body? by the fictional alien race The Krel
of them. Twenty years later, the first the film, a simulation replaced the phys- electrode arrays to brains. That’s dif-
“Star Wars” film showed something ical world. Life was lived entirely within ficult, but not impossible. In the next
vaguely similar: An eccescopic depic- cyberspace. In such a constructed world, 20 years, direct brain/computer inter-
tion of Princess Leia in a beam of light. nothing is real, yet anything becomes face technology is likely to advance far
Ten years after that, “Star Trek the possible. People can have superpowers, beyond what we can do today. Science
Next Generation” introduced the Ho- and objects can change form or even has already advanced considerably
lodeck, a completely immersive alter- disappear instantaneously. in this direction. No, the basic prob-
nate reality in which everything could Yet a direct-brain interface like lem is your perception of reality is
be eccescopic. the one in “The Matrix” turns out to already a construct—one maintained
by your brain. For example, you don’t ty to “paint” a 3-D shape onto individual
literally see things the way a camera dust particles floating in the air.
does. At any moment in time, your Whereas Holodust creates a glow-
eyes perceive only a tiny window into ing image of an object that seemingly
Distinguished reality from which your brain then
constructs a plausible model. It is re-
floats in thin air, the 360 degree Light
Field Display at USC is more ecce-
Speakers Program ally this constructed model that you scopic, because it allows the shading
“see.” We don’t know very much about of a virtual object to change as it is
how this construction process works, seen from different directions. Unfor-
http://dsp.acm.org which means we can’t hack into it tunately, that technology relies on a
with any effectiveness. And even if slanted metal mirror rotating at a very
we could, a direct brain interface like high speed. If you tried to touch it you
the one in “The Matrix” would need would most likely destroy both the dis-
to replace the considerable amount play and your hand.
of image processing done by our optic
nerve. We might also need to simulate THE FUTURE EVOLUTION OF
the saccades and other movements ECCESCOPIC TECHNOLOGY
made by our eyeballs as our brain con- Charles Darwin observed every geno-
Students and faculty tinually refocuses its attention. type requires a viable phenotype. That
The most reliable way to transmit is, no mutation can survive unless it
can take advantage of visual information to the brain is in can produce viable offspring. Technol-
ACM’s Distinguished the form of visible light. Why invent ogy is like biological evolution in that
something new, when you already it can’t just magically jump far ahead.
Speakers Program have something as powerful as the Every step along the path to innova-
human retina? tion needs to be useful, otherwise it
to invite renowned will die in the marketplace before en-
ECCESCOPIC PROTOTYPES abling the next step.
thought leaders in IN THE LAB For example, I don’t think we will
Around 2002, our research group at first achieve widespread eccescopy
academia, industry
NYU developed an early prototype of through surgery. Yes, technically we
and government Holodust—a kind of eccescopic display, could give everyone an artificial lens
which created virtual images of 3-D ob- implant, but until there is a good rea-
to deliver compelling jects directly onto a cloud of dust. Since son for such an intervention, people
you can never know the exact position won’t do it. It’s not even that invasive
and insightful talks of each particle in a cloud of dust, our eye surgery is so exotic. You probably
scheme used two scanning lasers: An know many people who have had cata-
on the most important infrared laser to sweep through the racts and are walking around today
topics in computing cloud looking for dust particles, as well with an acrylic lens implant or two.
as a visible laser along the same opti- You don’t know who they are, because
and IT today. cal path that could flash on command. it’s not something people generally
This dual laser approach gives the abili- talk about. The operation itself is
ACM covers the cost relatively simple and safe, requiring
only local anaesthetic and no stay in a
of transportation hospital. But it’s only done because it
for the speaker avoids blindness. A very different val-
ue proposition than, say, implanting
to travel to your event. People will become an artificial lens so you can do Google
used to seeing searches within your eyeball. Most
people won’t opt for invasive surgery
virtual objects unless it helps them to be more “nor-
superimposed mal,” however that word is currently
defined in their culture.
onto the physical Not too long ago, putting an elec-
world. Each of us tronic auditory enhancement device in
your ear was something you did surrep-
will have our own titiously. A hearing aid was something
personal view of this you tried to hide—ideally you didn’t
want anyone to know you needed one.
augmented reality. Recently there has been a fascinating
A
new virtual reality (VR) user reaches out and feels around in the dark, searching for the
controller yet unable to see where it is. Elsewhere, immersed in their own VR experience,
another user is oblivious to the other person who walks into the room; they are standing
side by side, but unaware of the other’s presence. Mixed reality approaches that allow
these real-world elements to be blended into VR can break the user’s sense of immersion in
their virtual experience. Our research on engagement-dependent mixed reality has solved this
problem, by selectively blending real elements into the virtual world as users wish to engage with
them, which creates a seamless interaction across the continuum of real and virtual reality.
The Oculus Rift’s Kickstarter cam- a number of sizeable problems in try- reality to be mapped to VR. However,
paign triggered a resurgence of in- ing to deliver a VR experience that is perhaps the most pressing concern
terest in VR head-mounted displays usable in the real “consumer” world. currently facing the VR community is
(HMDs). Advances in small form factor Some of these problems are funda- not related to the rendering or display
displays (e.g., the high refresh rate, low mentally technological. For example, of VR experiences, but instead how we
persistence, and high definition pan- the fidelity of the VR experiences in- interact with VR experiences and make
els typically used in mobile devices) creases—as our capability to render VR HMD usage compatible with real-
demonstrated high fidelity VR HMDs and display them does—through world environments.
were now not only technologically fea- more powerful GPUs, better displays,
sible, but a viable and affordable con- and wider field-of-view lenses. Simu- THE GLOVES COME OFF
sumer reality. What followed has seen lator sickness, another major prob- The problem immediately becomes
Image by Creativity103
the likes of Samsung (Gear VR), Sony lem for users, is being addressed with apparent when you first wear a VR
(Morpheus), HTC/Valve (Vive), Oculus/ additional sensing such as external HMD. You find yourself in a virtual
Facebook (DK1/2, CV1), and Google tracking cameras combined with world with no perception of reality.
(Cardboard) battling to be the leader in headset-based inertial motion sens- How do you interact with this new vir-
this VR renaissance. But there remains ing, allowing for every movement in tual world you inhabit? The go-to in-
Figure 4. When a person enters the same physical space as the VR user, they are faded into the virtual view. When the user
wishes to engage with them, they would become fully opaque. Left: reality; middle: low engagement; right: high engagement.
and incorporating reality, there is an- CONCLUSIONS: CAN I SEE YOU NOW?
other class of mixed-reality displays What we’ve described is a means to
currently in use. AR headsets typical- balance the needs of VR users who are
ly rely on a see-through display where torn between two worlds: reality and
virtual content can be rendered on virtuality. Users want to interact with
top of the user’s view of reality (as op- a virtual world, yet have to route said
posed to reality rendered virtually).
They occupy the AR part of the RV
interaction via the real world. This bal-
ance between immersion in virtuality ACM
continuum, with Microsoft Hololens and awareness of reality is key to pro-
being a recent example of the state of viding usable, yet immersive, VR head- Transactions on
the art. sets that can be used in homes, and
Because AR headsets have the abil-
ity to transition between a range of
offices, as well as other shared, social,
complex spaces without impediment.
Accessible
points on the RV continuum, we can
apply our engagement-dependent
Our work provides a foundation for
both AR and VR displays, with transi-
Computing
mixed-reality approach. This time tions in mixed reality governed by the
the blending would govern when and user’s engagement. This engagement-
how much virtual content is rendered dependent approach to mixed reality
on top of reality. For example, a lin- provides experiences that better adapt
gering gaze on a tourism landmark to user needs, and break down the
might be a cue to incorporate addi- isolating digital and physical barriers
tional relevant information. raised by head-mounted displays.
The underlying technologies used
in AR headsets suggest in the future References
there will be HMDs that support both [1] Wilson, G., Carter, C., Subramanian, S., and Brewster,
S. Perception of ultrasonic haptic feedback on
augmented reality and VR/augmented the hand: localisation and apparent motion. In
virtuality modes, and will have the Proceedings of the 32nd annual ACM conference on
Human factors in computing systems (CHI ‘14). ACM,
ability to transition between any point New York, 2014,1133-1142.
on the RV continuum. For example, we [2] Sodhi, R., Poupyrev, I., Glisson, M. and Israr., A.
can envisage Hololens–like headsets AIREAL: interactive tactile experiences in free air.
ACM Trans. Graph. 32, 4 (July 2013).
where there is an additional display [3] Simeone,A., Velloso, E., and Gellersen, H..
layer that can selectively occlude real- Substitutional Reality: Using the Physical
Environment to Design Virtual Reality Experiences.
ity, such that the rendering of virtual In Proceedings of the 33rd Annual ACM Conference
content can be done on top of reality on Human Factors in Computing Systems (CHI ‘15).
ACM, New York, 2015, 3307-3316.
or on top of a blank, dark canvas— ◆ ◆ ◆ ◆ ◆
[4] Machkovech, S. SteamVR: The room-scale VR
much like how 3-D TV active-shutter world that feels like an “IMAX in your house.” June This quarterly publication is a
glasses work today. Such headsets 13, 2015. Ars Technica UK. http://arstechnica.
would be at once empowering and iso-
co.uk/?p=19673. quarterly journal that publishes
[5] Milgram, P., and Fumio, K. A taxonomy of mixed
lating in equal measure, supporting reality visual displays. IEICE TRANSACTIONS on
refereed articles addressing issues
instantaneous transitions between Information and Systems 77.12 (1994), 1321-1329.
of computing as it impacts the
virtual spaces and augmented reality [6] Metzger, P.J. Adding reality to the virtual. In
lives of people with disabilities.
Proceedings of the 1993 IEEE Virtual Reality
spaces. But how would we make these Annual International Symposium (VRAIS ‘93). IEEE
transitions seamless? What elements Computer Society, Washington, DC, USA, 1993, 7-13. The journal will be of particular
of reality should get carried back and [7] McGill, M., Boland, D., Murray-Smith, R., and interest to SIGACCESS members
Brewster, S. A Dose of Reality: Overcoming Usability
forth? And how do we provide users Challenges in VR Head-Mounted Displays. In and delegates to its affiliated
Proceedings of the 33rd Annual ACM Conference on
with a stable mental model for inter- Human Factors in Computing Systems (CHI ‘15). conference (i.e., ASSETS), as well
acting with reality as their immersion ACM, New York, 2015, 2143-2152.
as other international accessibility
into VR increases? There would be
a need for consistent behaviors and Biographies
conferences.
rule sets regarding transitions across Daniel Boland applies data science and machine learning ◆ ◆ ◆ ◆ ◆
the RV continuum, be they pertaining approaches to HCI. This work was informed by his Ph.D.
thesis on adapting to user engagement. Learn more at www.acm.org/taccess
to interactions with reality, aware- www.dannyboland.com.
ness of social contexts, and so on. We Mark McGill is a Ph.D. student at the University of Glasgow
www.acm.org/subscribe
suggest an engagement-dependent in the Multimodal Interaction Group, led by Prof. Stephen
Brewster. Outside of VR, his research interests are
mixed-reality model could underpin in multi-user and multi-view TV. Learn more at www.
such interactions, and such a model is markmcgill.co.uk.
Managing
Cybersickness
in Virtual Reality
If the physical side effects associated with virtual reality are not
managed, the widespread adoption of VR may come to a halt.
By Lisa Rebenitsch
DOI: 10.1145/2810054
S
creenwriters, science fiction authors, you, and I… we all have dreamed of virtual
reality (VR) for decades. Hollywood whet our appetite for VR with films such as “The
Matrix,” “Total Recall,” and “Brainstorm.” As early as 1982, when “Tron” introduced
VR to the masses, the technology seemed well within reach for us all. VR was
present on the small screen as well as in the form of the Holodeck on “Star Trek: The Next
Generation.” By the early 1990s consumers could finally purchase VR gaming devices; there
was the Nintendo Virtual Boy, the Sega VR console, and the Sony Glasstron. The addition of
head tracking enabled dynamic VR applications, while 3-D screens heralded a new interface
mirroring the real world. Beyond entertainment, VR has also been used for military
and aviation training; its versatility lator sickness or visually induced mo- up to the software. Therefore, devel-
has been useful in design, data visual- tion sickness (VIMS), has been well opers simply needed to wait and even-
ization, and more. But just as VR began documented since the 1960–1970s. tually hardware improvements would
to peak, it nearly became extinct. These motion sickness–like symptoms eliminate cybersickness. There was
There are multiple reasons for this, were first reported while using military one major exception: In 1995, Mon-
but one of the most significant is the and aviation simulators. The term cy- Williams, Wann, and Rushton pre-
technology over-promised and under- bersickness is associated with full vir- dicted improving the hardware would
delivered. Military training simulators tual environments, while VIMS is used actually increase cybersickness [1]. To
were extremely expensive to operate. more broadly and refers to any situation better understand how cybersickness
Image by Durantelallera / shutterstock.com
Even as a commercial product, de- that involves looking at imagery, real or works requires delving into human
vices proved too costly for consumers. computer generated. Cybersickness is physiology and how we adjust to vi-
Despite implementing the most basic not the same thing as motion sickness. sual imagery.
technology, a monochromatic display, They differ with respect to their symp-
the Nintendo Virtual Boy was a failure. toms: Motion sickness is identified pri- WHAT IS CYBERSICKNESS?
There was also the issue of comfort. marily by nausea, while cybersickness There are a few common theories ex-
The Glasstron and Sega headsets were involves disorientation and dizziness plaining cybersickness, with the most
heavy. Worst of all, using VR devices usually preceded by nausea. prevalent being sensory mismatch.
made some people ill. The general assumption around This theory states if there is a mis-
Cybersickness, also known as simu- 1990 was the hardware would catch match between the sensory input to
VR IS BACK, BUT SO
IS CYBERSICKNESS
Nintendo Virtual Boy Oculus Rift
Recently, there have been significant
(a) (b)
changes to VR hardware: The costs
have come down and the hardware
cal quality with shadows and complex to experience symptoms. The results [1] Mon-Williams, M., Wann, J. P, and Rushton, S. Design
Factors in Stereoscopic Virtual-Reality Displays.
imagery are better tolerated. are preliminary since recruiting older Journal of the Society for Information Display 3, 4
Habituation. There is one thing that participants can be difficult. But un- (1995), 207-210.
is working in a developer’s favor—ha- like motion sickness, older users have [2] Graeber, D. A. and Stanney, K. M. Gender Differences
in Visual Induced Motion Sickness. In Proceedings
bituation. If someone uses an applica- shown a higher incidence of cybersick- of the Human Factors and Ergonomics Society 46th
Annual Meeting (Baltimore, September 30-October
tion regularly, they will slowly build up ness in some VR systems [4]. This may 4). Human Factors and Ergonomics Society, Santa
a tolerance. We are not sure if using a be an effect of video game usage cor- Monica, 2002, 2109-2113.
different VR application will give the relating to cybersickness, or it may be [3] Arns, L. L. and Cerne, M. M. The Relationship
between Age and Incidence of Cybersickness among
same effect, or how long it lasts, but something yet to be discovered. Immersive Environment Users. In Proceedings of
playing video games with substantial the IEEE Virtual Reality (Bonn, March 12-16). IEEE
Washington, DC, 2005, 267-268.
amounts of forward motion such as GUIDELINES FOR DEVELOPERS
[4] Kennedy, R. S., Lane, N. E., Berbaum, K. S., and
platformers and first-person shooters One of the best things developers can Lilienthal, M. G. Simulator Sickness Questionnaire:
does correlate with less cybersickness. do is include optional settings that An enhanced method for quantifying simulator
sickness. The International Journal of Aviation
Susceptibility. We have discussed decrease the likelihood of cybersick- Psychology 3, 3 (1993), 203-220.
hardware and software, but what ness. The easiest options are to shrink
about the person? Past studies suggest the field of view, slow forward speed,
Biography
20-40 percent of cybersickness can be give people at least some control over
Lisa R. Rebenitsch recently graduated with a Ph.D. in
attributed to the user alone. That is a their movement, and seat users. The computer science from Michigan State University.
great deal to ignore. Gender has been exception to seating people is that, in She joined the software engineering faculty at the
Milwaukee School of Engineering this fall. Her research
studied repeatedly. Many studies re- general, the more natural or antici- interests lie in cybersickness, virtual reality, and user
port women are more susceptible to pated the movement, the less cyber- interfaces in general.
cybersickness. However, gender also sickness there is. This means devices
correlates well with motion sickness such as the Cyberith Virtulizer and © 2015 ACM 1528-4972/15/09 $15.00
By Robert Konrad
DOI: 10.1145/2810048
Y
ou’ve probably never heard of the vergence-accommodation conflict, but if you’ve ever
felt nauseous watching a 3-D movie, you’ve felt its effects. Unfortunately, you might
face a similar problem with the current wave of virtual reality (VR) technologies.
Current 3-D movies and VR systems are based on planar stereoscopy, a process of
displaying two different images of the same scene on a planar screen from slightly different
angles—one to each eye. When viewed together, the images fuse and create an impression
of depth and solidity. While planar stereoscopy “tricks” our visual system into perceiving
depth, it also induces a conflict be- you can’t do one without the other. look and focus on the same object.
tween two parts of the human visual More importantly, the distances at However, there is a problem with ste-
system, causing some users to experi- which you converge and accommo- reoscopic displays.
ence pain and even nausea. This con- date are the same because you usually In these displays, each eye is pre-
flict is known as the vergence-accom- sented with a different image and
modation (VA) conflict. one is able to look at objects as if they
In order to understand the VA con- were in front of or behind the actual
flict let me first explain how our visual
system operates in the real world.
Possibly the 2-D display, depending on the dis-
parities for each object. This is great
When a person is looking at the world, most exciting because it gives us the impression of
their eyes respond in two ways. First,
the brain instructs the eyes to rotate
new approach in depth. However, our eyes must focus
on the display surface to keep the
so they are oriented toward the same creating a near images sharp on our retinas. See the
point of interest (this is called ver-
eye display with conflict? Our eyes converge on objects
Image by Alicia Kubista
gence). At the same time, the brain as if they were some distance away
forces the lens in each eye to focus so correct vergence and from the display, while they accom-
the point of interest appears sharp on
the retina. When looking around the
accommodation is modate on the display surface itself.
We are converging and accommodat-
world these two actions are coupled; light field displays. ing to points at different distances,
Biography
Robert Konrad is a Ph.D. student in electrical engineering
at Stanford University in Professor Gordon Wetzstein’s
group on computational imaging. He is currently focusing
on computational imaging and displays, as well as their
applications to virtual and augmented reality.
VR to the Rescue:
Avoiding danger
with mine disaster
simulations
Virtual reality is helping rescue teams prepare for emergency
situations in places they could never ordinarily go, like collapsed
mines deep underground.
By Alain Boulay
DOI: 10.1145/2810050
M
ining is a very dangerous profession. In addition to threats from falling rock deep
underground, there are poisonous gasses deep in the earth that can cause death
and explosive rock blasting hazards that cause unstable rock and mud slides, or
“muck.” Just recently, at Stobie Mine, a few young miners my age were killed due
to a muck slide, which occurs when underground water veins overflow and mix with crushed
rock. Safety is of paramount importance to the mining community, but it is challenging to
prepare for mine emergencies since real disasters cannot be contrived to practice rescue
operations in realistic conditions.
Mines have emergency rescue to work on, nor did I think my re- some well-worn paths for me to fol-
teams that go underground to res- search would lead me to developing low, so I started working on innova-
cue miners in the case of a disaster. virtual reality (VR) simulations for tions in 3-D VR and mining. The town
These emergency teams need to be emergency rescue missions in mines. where I study, Sudbury, is located in
Image by Przemek Tokar / shutterstock.com
trained to rapidly save lives without I hoped my background in psychol- northern Canada and is world famous
becoming victims too. Rescue teams ogy would help me develop insights for its nickel and copper mines. The
cannot train in real mine disasters into human-computer interaction re- mining industry’s influence is perva-
since it would be unethical to place search, but I was not sure what area I sive in the area; the industry helps to
them in dangerous situations. The wanted to specialize in. I now work in fund our university.
solution is simulations. This is where the Computer Human Interac- At MIRARCO Corporation, I have
3-D VR is important. tion Lab at Laurentian Univer- been developing 3-D mine rescue
Like many of you, when I started sity, which primarily focuses simulations that can be viewed on
my master’s degree in computational on brain-computer interfacing and multiple types of VR training plat-
science, I had no idea what I wanted 3-D VR. My predecessors provided forms. These platforms include 3-D
Figure 1. The iTX multi-gas monitor is widely used in the mining industry to detect
a variety of dangerous gases.
ACM
ACM Conference
Conference
Proceedings
Proceedings
Now
Now Available via
Available via
Print-on-Demand!
Print-on-Demand!
Visit us at
http://ubiquity.acm.org/blog/
PROFILE DEPARTMENT EDITOR, ADRIAN SCOICĂ
PROFILE
up during the Meiji era by Dr. Satō in a Im ilit wiscin henim zzriustrud eum eum eugait ex eu(MEL),”
Laboratory facin velent
Tachiadipit,
recalled.quip et
deeply academic home atmosphere, them as they please.
iuscipit, sed ero diamcon sequamet eriustie
After delisi. Ignimstudy,
a preliminary quissimthe project
was conducted on a six-year plan about ways to further advance his work. He rejoined the University of Tokyo as
that established technology related His aim was to create a simulator device a professor in 1989 and has continued
to the intelligent navigation of mobile by which humans could work freely in his research in telexistence with more
robots, such as routing with the use an environment with obstacles without than a hundred research members
of landmarks and obstacle detection. putting themselves in danger, which in and students in the Tachi Laboratory.
This work led to the development of turn led him to conceive the idea Since he believes only experience
the idea of intelligent disobedience. of telexistence. can make people understand the
Tachi continued, “Right now there “I came up with the idea of true meaning of telexistence, he has
are very good navigation systems, telexistence in the early autumn of constructed TELESAR (TELExistence
but back then there weren’t any, so that year. When walking down the Surrogate Anthropomorphic Robot) to
we had registered landmarks and corridor, I was suddenly reminded demonstrate the advancement
searched for a route. Sometimes, that all of human vision is just of the technology. He has invented
it might be very dangerous for a based on two images projected on several innovative technologies in 3-D,
person to walk straight due to various our retinas, and we build the 3-D virtual reality, augmented reality,
kinds of obstacles, at which point we world by moving our heads and and haptics, which have in turn
realized that it would be better for exploring the world. So if we replace advanced telexistence.
the guide dog robot to disobey the that with a virtual image, what we
commands of the master and thus, to experience is a kind of virtual world. NEW DIRECTIONS IN TELEXISTENCE
protect him. We called that intelligent Since we perceive the environment According to Tachi, building technology
disobedience.” through sensors and reconstruct that is capable of telexistence will have
According to Tachi, the project the world in our brains, if we gather a definite positive impact on our rapidly
had other important outcomes, and that information from a robot and aging, transportation-reliant society: “If
in particular it confronted them with present it to the human, we can we could succeed in transmitting haptic
thinking about the problem of human- live through the experience of sensations, then society will change, so
machine interaction, as there was being embodied within the robot, that real work can be done remotely, in
a need for the guide dog robot and existing in its specific environment. environments which would otherwise be
the human to constantly exchange I immediately went back to my office hazardous or hard to reach. We could do
information about the environment. and wrote ever ything down, and international timeshare, and we could
simultaneously came up with the idea allow disabled or otherwise old people to
VISIT TO MIT, AND THE of how to design the visual display to continue working. In Japan, for example,
BEGINNINGS OF TELEXISTENCE do that. 1980 was the beginning of we have a very aged society, but in many
In 1979, Tachi joined the famous telexistence,” Tachi explained. cases old people can still work, and would
Professor Robert W. Mann’s lab at Despite the simplicity and elegance like to keep contributing to society even
MIT as a Japanese Government Award of his revelation, Tachi confessed his if they get somehow physically impaired
Senior Visiting Scientist in order to do ideas regarding telexistence were at with age. They have lifelong experience,
joint research on the guide dog project. first met with skepticism, and only and this experience is very valuable.
He felt privileged to work at the same after developing a prototype did his Telexistence would liberate them from old
university where Wiener had initiated peers truly understand the out-of- age, and maybe even change the way
cybernetics. After what turned out to body experience he had been talking we socialize.”
be an extremely productive year, he about. “I was immediately impressed On a final note, Tachi remarked the
returned to Japan in 1980 thinking when I saw myself for the first time holy grail of telexistence is to have
using the system I had built. It’s robots connected to the Internet and
not like a mirror. When I saw myself spread all over the globe, with people
from the third person using that able to log on and off to them as they
system, I began questioning whose please. He emphasized telexistence is
There was no consciousness I really was, allowing
me to perceive my body like that,” he
in a sense the ultimate form of cyborg:
“Using this technology, human beings
computer science said. Tachi also observed in the age can obtain an augmented superhuman
How to apply:
Online: https://application.heidelberg-laureate-forum.org/
Materials to complete applications are listed on the site.
Stanford, California
master’s students and a large group of
undergraduate research assistants—so
it’s a busy place. I feel really grateful to
be part of such an amazing group. The
graduate students in the department
Editor’s Note: In this issue, Andrea tual reality in head-mounted displays were helpful to me when I first started
Image by Linda A. Cicero / Standford News Service
Stevenson Won shares her experience at like the Oculus Rift. Our research in- and I hope to be able to pass that along
Stanford University and her pursuit of cludes tracking and representing hu- to newer students. I love collaborating
research in virtual reality to help others. man behavior. There are three major with my fellow students; it is fun to be
—Somdip Dey areas of investigation. First, how can around other people who get so excited
T
virtual reality be used to study hu- about experimental results. There are
he Virtual Human Interaction man behavior in the real world? Sec- also lots of opportunities to collaborate
Lab (VHIL) at Stanford Uni- ond, how can virtual reality be used with people from other departments,
versity uses immersive virtual to positively affect the real world? for example, education, psychology,
environments to study human Third—and this is becoming increas- political science, and medicine. Our re-
interactions. However, the lab’s re- ingly relevant as these kinds of sys- search is very cross-disciplinary.
search is not just limited to classic vir- tems become more widely used—how When I started at Stanford I was im-
VR Head-Mounted
work. This made for a pretty intense
first year, but it was what I was look-
ing for in a lab. My advisor, Jeremy
Bailenson, has a reputation for being
very productive, which means his advi- Displays
sees have a lot of opportunities to get
involved with interesting projects. It’s hard to ignore the proliferation of head-mounted displays lately.
Before coming to Stanford my back- Google Glass, Microsoft Hololens, and Oculus Rift are all well-known
ground was in anatomical models, fo- products that have entered, or are poised, to enter the market. This
cusing on creating digital models for isn’t the first time we’ve seen a frenzy of interest in augmented and
virtual reality. My M.S. from the Uni- virtual reality (VR) though. By the early 1990s, VR was already a hot
versity of Illinois at Chicago focused topic in mainstream media and a handful of early consumer products
on biomedical visualization, and I did enjoyed limited success.
a lot of custom work with clinicians VR games had already reached arcades by the early ‘90s, and in
after I graduated. I became interested 1995, Forte Technologies released one of the earliest VR devices for
in how custom models like these could the home PC. Due to its relatively competitive price point, and modest
be used therapeutically, for exam- hardware requirements, the VFX-1 was one of the more successful
ple, showing patients suffering from offers on the market. Sporting the “cyberpuck” controller with built-
chronic pain a virtual representation in motion sensor, the VFX-1 could be used to play numerous games.
of their own face being painlessly Nevertheless, the VFX-1 suffered from technical shortcomings and a
touched. I became fascinated with generally poor user experience that fated it to limited adoption.
the idea that changing how a person In 2012, Oculus VR launched a wildly successful Kickstarter
is represented in media could change campaign to help fund further development of its product, the Rift,
their life in the real world. This idea quickly surpassing its funding goal. Developer Kit 1 began shipping
led me to pursue my Ph.D. at VHIL. in early 2013, the much improved Developer Kit 2 was released in
My main research interest is using 2014, and the consumer version is set to be released in 2016. With
virtual reality therapeutically for pain a wider field-of-view, better resolution, better tracking, and overall
patients, by changing how their bodies improved technology, the Rift promises to be a much more immersive
appear to them when they view them- experience than any VR hardware that has come before.
selves in a virtual reality. However, I am Despite the shortcomings of products in the past, this time really
broadly interested in how people inter- does feel different. Perhaps we will all finally get to experience
pret the physical world based on cues movies, games, and other digital media in the new and immersive
they receive through media, whether it ways that we’ve been promised for so many years.
Image of VFX-1 by eVRydayVR. Image of Oculus Rift Dev Kit 2 by Ats Kurvet.
E
very superhero has its page with the relevant information
own unique set of skills. about Thor’s known abilities. We then Figure 1: Part of a Wikipedia page
Just about any Iron Man- computed the semantic similarity showing Thor’s abilities.
obsessed fan could tell you between texts of different abilities
that. They could also give you an to quantify how similar the abilities
in-depth description of any of their of different superheroes are. Finally,
favorite comic book characters and we constructed the network of comic
respective superpowers. Well, we have book characters; two characters
news for you and that obsessed fan; were linked if their abilities matched
there are many more characters in by at least 80 percent according to
the Marvel Universe than you might Levenshtein-based similarity.
have thought, and most of your Stan Lee ran out of ideas 641
favorite childhood heroes possess superheroes ago.
powers similar to those of Stan Lee’s When talking about network
superheroes. properties, one must first focus on
As it turns out, there are more the number of nodes and edges in
than 900 characters hidden within the the network. Our network has 752
Wikipedia category “Marvel Comics nodes, with each node representing a
superheroes,” and approximately different Marvel superhero. There are
85 percent of entries include a 9,496 edges connecting the nodes,
description of superpowers right at where each edge indicates characters
the top of the page. This information who share many superpowers.
was the basis of our analysis. The network has 18 connected
In this column, we present the components of which the largest
uncanny resemblance between the component has 641 nodes. Next, we
superpowers of your favorite bedtime focused on the analysis of the largest
story companions and the vast connected component.
superhero network provided by the Each character in the network is
Marvel Universe and its inhabitants. on average linked to 30 superheroes.
Based on the definition of our network,
SCRAPING WIKIPEDIA the character with the highest
To obtain a list of Marvel Comics degree has the most “typical” skill
characters, we used a Python library set. Shatterstar, a mutant hero and
Wikipedia (http://github.com/ member of the team X-Force, is a
goldsmith/Wikipedia), which gives character with the most connections
us access to Wikipedia data that (see Figure 2). He is closely followed
can be parsed. The Wikipedia library by Ikaris with degrees of 132 and 128,
wraps the MediaWiki API; this is a respectively. In accordance with this
web service that provides convenient information, the most generic super
access to article summaries, wiki powers include superhuman strength, Many pages, such as this one, were used
features, meta-data, and data, such speed, agility, healing power, energy to extract information for our network
as links and images from Wikipedia manipulation, and enhanced cognition. analysis. We considered word tokens
describing the hero’s abilities (see
pages. After we identified the relevant The longest shortest path, i.e. “Abilities” in the figure) to quantify how
characters, we retrieved their the network diameter, is 15. similar two heroes are to one another
by counting the minimum number of
Wikipedia pages and extracted the Aragorn and Apex have the highest
operations required to transform the text
text referencing abilities associated eccentricity scores and are the on abilities of one hero into the text on
with each character. Figure 1 shows furthest apart in our network. This abilities of the other.
an example of a portion of a Wikipedia makes sense because their skill sets
Figure 3 shows a network centered [6] Stiller, J., Nettle, D., and Dunbar, R. I. The small world
of Shakespeare’s plays. Human Nature 14,4 (2003),
on Spider Man, Thor, Iron Man, and 397-408.
Captain America. According to the
large number of edges connecting Biographies
Spider Man and Thor, the two appear Lara Zupan is a high school junior attending Gimnazija
to be the most alike in regards to skill ROLLOUT Vic, Ljubljana, Slovenia. She has edited several high
school publications, designed Novi Dijak, a student run
set. They are both supernaturally In recent years, the increasing magazine targeting Slovenian high school students,
strong and experience superhuman availability of digitized fiction and is a member of the European Youth Parliament.
She is interested in graphics design, literature, and
longevity. The superhero with the and literary works has enabled visual arts.
most distinct skill set in this group is computational analysis of these Marinka Zitnik is a Ph.D. student in computer science
at the University of Ljubljana. She also did her
Captain America, with only one edge texts. These approaches have often research at the University of Toronto, Imperial College
connecting him to the rest of the been used together with more London, Baylor College of Medicine and Stanford
University. Her interests include machine learning,
superheroes. Iron Man is most similar traditional techniques of literature artificial intelligence, probabilistic numerics, and
to Spider Man as they are connected analysis. Recently, ideas coming from bioinformatics.
by a few edges, and they both have the analysis of complex networks
genius-level intellect. offer new ways to understand how Copyright held by Owner(s)/Author(s).
BEMUSEMENT
http://xkcd.com/1505/
Psychic
http://xkcd.com/628/
SUBMIT A PUZZLE
Can you do better?
Bemusements would like your puzzles and mathematical games
(but not Sudoku). Contact xrds@acm.org to submit yours!
EXTRAORDINARY WORK
Inside our walls, you will find the most extraordinary people doing Computer/Electrical Engineering
the most extraordinary work. Yet it’s not just finite field theory, Computer Science
quantum computing or RF engineering. It’s not just discrete Cybersecurity
mathematics or graph analytics. Information Assurance
Mathematics
Foreign Language
It’s all of these and more – rolled up into an organization that leads Intelligence Analysis
the world in signals intelligence and information assurance. Cryptanalysis
Inside our walls you will find extraordinary people, doing Signals Analysis
extraordinary work, for an extraordinary cause: Business Management
Finance & Accounting
The safety and security of the United States of America. Paid Internships, Scholarships & Co-op
U.S. citizenship is required for all applicants. NSA is an Equal Opportunity Employer and abides by applicable employment laws and
regulations. All applicants for employment are considered without regard to age, color, disability, genetic information, national origin, race,
APPLY TODAY
religion, sex, sexual orientation, marital status, or status as a parent.