Virtual Reality & Software Technology, 8/94 Singapore --[how these two subjects

[David Farber <farber () central cis upenn edu>]

 Dr. David K. Kahaner
 US Office of Naval Research Asia
 (From outside US):  23-17, 7-chome, Roppongi, Minato-ku, Tokyo 106 Japan
 (From within  US):  Unit 45002, APO AP 96337-0007
  Tel: +81 3 3401-8924, Fax: +81 3 3403-9670
  Email: kahaner () cs titech ac jp
Re: Virtual Reality & Software Technology, 8/94 Singapore
10/04/04 (MM/DD/YY)
This file is named "vrst.94"


ABSTRACT. Summary of the conference, Virtual Reality & Software
Technology (VRST'94), held at the Institute for Systems Science (ISS),
Singapore, 22-26 August 1994, is presented. This report contains text
and general comments on the conference. An accompanying report
(vrst-abs.94) contains abstracts of the papers that were presented.


This report was prepared jointly with Dr. Thomas Davis of the USAROSR
Tokyo office (Email: daviste () emh yokota af mil). I am also grateful for
comments from Dr. Tim Poston at ISS (Email: tim () iss nus sg) and Dr Hugh
L. Applewhite of Piltdown Inc, (Email: hugha () teleport com), which have
been included.


Virtual Reality Software and Technology '94


Event Organization and Background.


Virtual Reality Software and Technology '94 (VRST '94), which is
intended as the first in a series of annual VR conferences, was
conducted at the Institute of Systems Science (ISS) of the National
University of Singapore on 23-26 August, 1994. VRST '94 was co-sponsored
by ISS and the Association for Computing Machinery's Special Interest
Group for Computer-Human Interaction (ACM/SIGCHI). The General Chair for
the conference was Dr. Juzar Motiwalla, Director ISS. His contact
information, along with that of Dr. Steve Feiner from Columbia
University, who served as co-chair of the Program Committee, appears
below.


        Dr. Juzar Motiwalla
        Director, Institute of Systems Science
        National University of Singapore
        Heng Mui Keng Terrace,  Kent Ridge, Singapore 0511
         Tel: +65 772-2075; Fax: +65 775-0938
         Email: JUZAR () ISS NUS SG


        Dr. Steve Feiner
        Associate Professor, Department of Computer Science
        Columbia University, New York, NY 10027
         Tel: (212) 939-7083
         Email: FEINER () CS COLUMBIA EDU




The VRST '94 technical program is summarized below. As noted there, the
first day of the event (Tuesday morning, 23 August) was devoted to a
collection of parallel tutorial sessions. The primary conference
technical program began the following morning (Wednesday, 24 August)
with the program co-chair opening remarks, guest speech and keynote
address. The remainder of the conference technical program consisted of
paper, demonstration/video and panel sessions. Abstracts of all
presented papers are also attached along with as author contact
information.  For copies of papers contact the authors directly or Dr
Motiwalla at ISS. A Proceedings of the conference was published by


    World Scientific Publishing
    PO Box 128, Farrer Road, Singapore 9128
      ISBN 981-02-1867-2
   (G.Singh, S.Feiner, D.Thalmann, ed)


About 120 conference registrants distributed as follows.
      Singapore      67
      Japan          11
      USA             9
      UK              7
      Germany         5
      Korea           4
      Sweden          2
      Canada          2
      India           2
      Netherlands     1
      Switzerland     1
      Belgium         1
      Thailand        1
      Spain           1
      Austria         1


Participation would have been greater, but a large IEEE conference was
also being held in Singapore at the same time.


As noted above, VRST '94 is intended as the first in a series of annual
virtual reality workshops. During the opening session, Tokyo and
Switzerland were announced as the venues for VRST '95 and VRST '96
respectively, however no specific dates were established for either
event.


> From Dr Hugh L. Applewhite...
Piltdown, Inc.
4470 SW Hall Blvd, M/S 154
Beaverton, OR 97005 USA
  Tel: 503.641.8575, Email: hugha () teleport com


In general, it was a productive meeting.  There were enough people
representing sufficiently diverse areas that it wasn't three days of
esoterica, but rather a general sweep across the set of VR disciplines.
And there were experts in most of the areas, so that it didn't turn into
a 'wouldn't it be nice if ...' gabfest, unburdened by any facts or
figures.  A nice balance, I think.


Position tracking, a subject of perennial complaint, played a minor role.
The practical reason for this is that most VR applications have been
designed to take place within quite restricted working volumes.  Short
cables and interaction metaphors such as walk-in-place (Slater, Usoh &
Steed) and hand gestures (Wexelblat; Quek) ensure that existing magnetic
and ultrasonic trackers are adequate to the task.  Nevertheless, many
participants wistfully looked forward to the development of more capable
large-volume trackers.


Particularly interesting was the widespread agreement between both
application developers and tracker designers that the 'ultimate' tracker
is a combination of a short-range, accurate technique (e.g.  ultrasonic)
and a large-volume, untethered system (inertial).  Two tracker designers
presented systems: E. Foxlin *showed* a relatively small (7 cm cube) that
is light enough for head tracking.  H. Applewhite presented a new
ultrasonic phase coherent approach that promises absolute positioning as
well as high data rates.




> From Prof Timothy Poston....
Institute of Systems Science
National University of Singapore
Heng Mui Keng Terrace, Kent Ridge, Singapore 0511
 Email: TIM () ISS NUS SG


As its title suggests, VR Software and Technology '94 was concentrated
strongly on concrete work towards bringing out the goals of VR, rather
than discussion of its psychological aspects.  Regrettably but
unsurprisingly, this produced a heavy gender imbalance; of the two
scheduled talks by women, the more technical (on a fuzzy logic analysis
of human motion) was pre-empted by a honeymoon, and delivered by a male
stand-in.  The other concerned schemata for interactive drama, in general
as well as in VR.


The other imbalance common in VR meetings, between Asia and the North
Atlantic, was much better addressed by the siting of VRST'94 in
Singapore.  About a third of the presentations (counting either by
country of origin, or location of lab) were from Asia, in contrast to the
very thin Asian presence common in US meetings in VR.  The single
presentation of VR as an already-functioning element of commerce, in
fact, was Nomura's talk on a kitchen (and now house) planning system.  A
major point of interest there was the retreat from the use of a head
mounted display, and even from stereo: the new system uses a curved semi-
surround screen, big enough that the problems of coexistence in a virtual
space -- much discussed at the meeting -- can be solved simply by putting
married potential customers together in the middle of it.  This lacks the
futuristic appeal of the total-immersionist approaches suggested from
experimental laboratories, but it will make serious money for Matsushita,
in a practical context.  The same will not be true of head mounted
immersion until the field of view is that of the archers who won the
battle of Agincourt, rather than the narrow helmet slot of the French
knights who lost it.  (Could there be a game here?  Jousting is the only
sport where real vision is blinkered enough to be easily modelled in a
1994 electronic helmet.)


Much of the work presented involved particular 3D interfaces, but none
addressed the general issues of 3D interface design; how should the
choice of tools and actions be made clean, intuitive, and unfatiguing in
long sessions?  It may be that through the current helmet slot, no
interface can achieve this; at a conference that included a system for
constructing behavioral circuits by walking yards between the nodes, it
was easy to doubt that the VR community would yet recognize an interface
that did.


For me the most interesting talks were those that addressed the myriad
technical problems of VR, whether matters that should be handled
transparently (like tracking or collision detection, which the user only
notices when they fail) or how very specific tasks in virtual space are
being achieved -- the star here being Steve Bryson's boom-displayed
virtual wind tunnel at NASA.  This uses resources that other participants
can only dream of (Cray computation, a gigabyte of RAM, one Reality
Engine per eye for the display...), pressing the envelope everywhere with
tightly optimized code, and achieves results not possible without the
virtual environment.  That must be the real test of VR; mere immersion in
an unreal environment can be achieved far more cheaply by chemical means
or anaerobic exercise.  "The point, however, is to change it."


[Poston and colleagues' work on "reach in" VR technologies might be of
particular interest in areas related to precision manufacturing. I felt
that this was one of the more interesting original contributions from
Singapore at this conference, DKK.]




Technical Program Synopsis.


The VRST '94 technical program included a full day (Tuesday, 23 August)
devoted to virtual reality tutorial sessions. The tutorials were
conducted in parallel with one full day and four one half day programs.
The titles, instructors and durations of the tutorial sessions are
listed below. Tutorial number 2, titled Three Dimensional Graphics
Hardware, was cancelled at the conference site, but all others were
conducted as scheduled.


Tutorial 1 (Full Day): Introduction to Virtual Reality Design


Steve Bryson, NASA Ames, USA
Steven Feiner, Columbia University, USA
Tutorial 2 (Half Day): Three Dimensional Graphics Hardware (Cancelled)
Tai-cker Chiueh, State University of New York, Stony Brook, USA


Tutorial 3 (Half Day): dVise - Using the Virtual Design Environment(tm)
with your Existing CAD Models
Pierre duPont, Division Limited, UK


Tutorial 4 (Half Day): Geometrical Modeling and Animation for VR
Mark Green, University of Alberta, Canada


Tutorial 5 (Half Day): Graphical and Acoustical Rendering for VR
Peter Astheimer and Stephan Muller, Fraunhofer-Institute for
Computer Graphics, Germany


The full day tutorial, titled Introduction to Virtual Reality Design and
conducted by Bryson and Feiner, was a very comprehensive survey of the
field. It included sessions on interface technology, human factors,
performance issues, architectures and case studies. The emphasis
throught was on state of the art developments. The course handout
material consists of some 84 pages of viewgraph hardcopy (two per page).


The main conference program began on Wednesday morning, 24 August with an
opening speech by Radm. Teo Chee Hean, Singapore Minister of State for
Finance and Communications. Adm. Hean addressed the conference in the
context of virtual reality as a component of Singapore's Information
Technology national research and development objectives. He indicated
that Singapore considers information technology to be a key to its long
term competitiveness and intends to vigorously pursue international
linkage opportunities such as VRST '94 in the future. The Minister cited
several examples of the use of VR in Singapore, including architecural
walkthroughs of the Performing Arts Center using WaveVisions, a product
partially developed at the National Computer Board's Information
Technology Institute and also via collaborations with ISS. He also cited
use of VR at the Singapore General Hospital for  surgery preparation, at
the Ministry of Defence and the Port of Singapore for training of
military personnel and operations staff such as crane operators.
Repeatedly, I have heard Singaporean speakers emphasize that the country
cannot afford to devote its limited resources to opportunities that will
not offer significant payback, and the Minister made the same comments
here. He specifically mentioned five thrusts related to Information
Technology R&D, and I have attached those remarks below.


"First, anchor IT R&D to support IT2000. IT2000 is the IT response to the
Government's vision articulated in "The Next Lap". IT2000 aims to
transform Singapore into an Intelligent Island where IT is pervasively
used in both businesses and homes. Business competitiveness would be
greatly enhanced by IT and so will the quality of life enjoyed by every
Singaporeen. IT R&D is important to help make this vision a reality.
Many technological breakthroughs have already been made in laboratories
worldwide. The challenge to bring forth a new generation of leading IT
applications and break this bottleneck from laboratories to real
applications must be made a priority. The IT2000 effort will offer
tremendous opportunities to experiment and use new technologies for
advanced applications.


"Second, focus on market driven IR R&D. IT R&D efforts must co-evolvewith
application needs. There is increasing pressure favouring application
driven R&D. The early experience of lead users of new R&D ideas can help
guide and influence research direction and new product.innovation. In the
IT2000 study carried out by the National ComputerBoard, over 60 strategic
sector wide applications were identified to help support our nation
building efforts. One example is CORENET to link the construction and real
estate sector in Singapore. This collection of sector-wide driven
applications can provide a good starting point for identifying impactful
projects to spur national development. Apart from contributing to national
competitiveness and better quality of life,these applications can also be
resold and contribute to national wealth.


"Third, Intensify efforts to develop new competencies. Certain key
technologies may dominate the industry. When such critical technologies
are identified, new programs must be quickly developed to build up core
competencies in them. For example, the digital media revolution is
changing the IT Industry as we know it today. It is therefore imperative
that new initiatives are quickly implemented to build competency in
exploiting digital technologies. I am pleased to note that as a start, a
Singapore Digital Media Consortium, comprising four local companies,
namely, Aztech Systems, Creative Technology, IPC and iMedia Singapore,and
three local research institutes, has already been formed to work on
upstream digital media R&D. Research collaborators will include the
renowned Media Lab from the Massachusetts Institute of Technology.


"Fourth, Invest In training. In atiticipation of the Industry needs for
the new digital media industry and information superhighways, more
preactive training must be developed and introduced. The tertiary
institutions must ensure that new graduates are equipped with the core
skills needed to support advanced IT innovations. Equally important, the
pool of existing IT professionals must be continually re-trained in new
technologies and be kept abreast of IT-developments worldwide.


"Finally, capltalize on international linkages. As a small nation, we do
not have the resources to effectively undertake all the IT R&Dchallenges
identified. Therefore, we must work with intertiationalpartners and
leverage on their capabilities and resources to helpsupport our IT
aspirations. The collaboration with internationallyrenowned R&D centres
would help transfer much needed skills andexpertise to the country.
Singapore also adopts a liberal policy wheretalented individuals from
abroad can easily settle here to participatein high value added
activities.


"I am glad to note that the National Computer Board and the National
Science and Technology Board are working closely together to implement
these thrusts and create a vibrant IT R&D environment in Singapore. I am
told that discussions are already underway to implement many of the
initiatives identified.


"The implementation of the IT2000 master plan would generate many
opportunities for new technologies to be used. Virtual Realitytechnology
is already making an impact here and I am sure the demand forit will
continue to grow. To all VR evangelists who are attending thisconference,
I hope that you will have useful discussions and find newways of using
the technology meaningfully."


The next event on the conference agenda was the keynote address (an
abstract is attached below), delivered by Michael Deering from Sun
Microsystems. A central theme of his address was a concern that the VR
research communuity might allow expectations to exceed what can be
delivered. He noted that about a decade ago the Artifical Intelligence
community allowed something of the sort to happen to its discipline,
with adverse consequences. Other key points in his address were that (1)
successful commercialization of VR will require somethingh on the order
of a $10K system which provides about 1.5 times productivity improvement
at about 2hrs/day utilization and allows a 2 year capital recovery
period, and (2) that the VR research community needs to quickly answer
the question "What is the killer application?"


The remainder of the conference technical program was divided into a
series of paper, panel and demonstration/video sessions. Abstracts of
all presented papers, demonstrations/videos and panel session opening
remarks, collected by session title are available in the companion
report "vrst-abs.94".