ICAT '93 - the third International Conference on Artificial reality and Tele-existence July 6-7, 1993, Tokyo, Japan.

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

From: 
 Dr. David K. Kahaner

ABSTRACT. ICAT '93 - the third International Conference on Artificial
reality and Tele-existence July 6-7, 1993, Tokyo, Japan.


Report on ICAT '93 - the third International Conference on
          Artificial reality and Tele-existence
              July 6-7, 1993, Tokyo, Japan

Submitted by: Dr Mei Kobayashi ( LAB-S73 )
              Media Systems Institute, IBM Tokyo Research Laboratory
              1623-14 Shimotsuruma, Yamato-shi, Kanagawa-ken 242 Japan
               Tel: +81 462-73-4934;  Fax: 81+462-73-7428
               Email: MEI () TRLVM VNET IBM COM

The following is a report on ICAT '93, the Third International
Conference on Artificial reality and Tele-existence held July 6-7, 1993
in Tokyo, Japan. Readers are forewarned that this article has been
written by a someone who did not actually attend the meeting; the report
was written after discussions with a limited number of attendees.  A
second note of caution: this is an English report on Japanese reports
about a workshop held in English here in Japan. In short, facts and
details about the workshop have been passed through several unusual
filters so that the output may be slighly distorted or skewed in
perspective. Many thanks to Dr. David Kahaner for his kind words of
encouragement, "epsilon.gt.0"; indeed I hope that this limited report is
better than none. [Readers may want to reveiw last year's conference, 
"icat.92", 5 Aug 1992, DKK.]

The report consists of four sections:

1. Administrative Information and Conference Program
2. ICAT '93 Proceedings: Table of Contents
3. ICAT '93 Video ( review and comparison with
                    the video:  Virtual Reality '93 (JTTAS) )
4. Impressions and Tidbits on the Conference

Official papers and documentation will be covered in parts 1 and 2.  I
will pretty much limit discussion in parts 3 and 4 to impressions and
trends sensed by attendees rather than technical details. Interested
readers are requested to directly contact speakers and/or the organizing
committee regarding purchases of the proceedings, technical content,
details, etc..

________________________________________________________________________

1. Administrative Information and Conference Program

Sponsors: Japan Technology Transfer Association (JTTAS)
             2-11-2 Nagata-cho, Chiyoda-ku, Tokyo 100 Japan
             tel: 03-3597-8220, fax: 3597-8224
          The Society of Instrument and Control Engineers
          Nihon Keizai Shimbun, Inc.

Organizing Chair: Prof. Susumu Tachi, RCAST, U. of Tokyo
          Research Center for Advanced Science and Technology (RCAST)

Organizing Co-Chair: Prof. Michitaka Hirose, The Univ. of Tokyo

International Organizing Commitee:
Steve Ellis, NASA Ames Research Center
Yukio Fukui, Agency of Industrial Science and Technology (AIST), MITI
Hiroshi Harashima, The Univ. of Tokyo
Toyohiko Hatada, Tokyo Inst. Polytechnics
Katsura Hattori, Asahi Shimbun
Takayuki Itoh, NHK
Hiroo Iwata, The Univ. of Tsukuba
Robert Jacobson, World Design, Inc.
Myron W. Krueger, Artificial Reality Corp.
Carl Loeffler, Carnegie Mellon Univ.
Taro Maeda, The Unov. of Tokyo
Ryo Mochizuki, Media Int'l Corp.
Warren Robinett, the Univ. of N. Carolina
Makoto Sato, Tokyo Inst. of Technology
Thomas B. Sheridan, MIT
Robert Stone, Nat'l Advanced Robotics
Gen Suzuki, NTT Human Interface Laboratory
Haruo Takemura, ATR Communication Systems Research Laboratory
David Traub, Compec

I have not listed the program since it is a subset of the table of
contents of the Proceedings given below. The conference consisted of two
days of talks, demos and a panel discussion with lunch and coffee breaks
breaks at comfortably spaced intervals. Keynote lectures by Loeffler
(7/6), Harashima (7/7)  and Ellis (7/7) kicked off both days of the
meeting. General talks were gievn in three technical sessions:
1. Virtual Reality Communication and Control, 2. Virtual Environment
Construction, 3. Force Display and Shape Modeling. In addition to the
technical program, there was the first Inter Collegiate Virtual Reality
Contest (ICVRC '93) and a Beer Party on the evening of the 6th.
A high registration fee and length ( 2 weekdays ) were enough to make
corporate managers think carefully about the number of researchers to
send to the meeting.  One estimate put the number of attendees at
somewhat over 100.

________________________________________________________________________

2. ICAT '93 Proceedings: Table of Contents

Marks:
   K = Keynote Speech
   1 = Technical Session I: Virtual Reality Communication and Control
   2 = Technical Session II: Virtual Environment Construction
   3 = Technical Session III: Force Display and Shape Modeling
   O = Other Papers without Presentation

K  Carl Loeffler ( Project Director, Carnegie Mellon U. )
      Distributed Virtual Reality: Applications for Education,
      Entertainment and Industry

1  Susumu Tachi, Kenichi Yasuda ( RCAST, U. of Tokyo )
      Evaluation Experiments of Tele-Existence Manipulation System

1  Haruo Takemura, Yasuichi Kitamura, Fumio Kishino, Jun Ohya
      ( ATR Communications Systems Research Laboratory )
      Distributed Processing Architecture for Virtual Space
      Teleconferencing System

1  Michael Zyda, Chuck Lombardo, David R. Pratt
      ( Naval Postgraduate School ) Hypermedia and Networking
      in the Development of Large-Scale Virtual Environments

2  Ryugo Kijima, Michitaka Hirose ( U. of Tokyo )
      Virtual Sand Box: A Development of an Application of Virtual
      Environment for the Clinical Medicine

2  Gen Suzuki, Shouhei Sugawara, Machio Moriuchi
      ( NTT Human INetrface Laboratories ) Visual Communication
      Environment Using Virtual Space Technology

2  Kazuo Itoh ( VR Project Manager, Asahi Electronics, Inc. )
      PC Based VR Systems

K  Hiroshi Harashima ( U. of Tokyo ) Face Expression and Communication

K  Steve Ellis ( Head: Spatial Perception and Advaced Display Lab.,
      NASA Ames Research Center ) What are Virtual Environments ?

3  Hiroo Iwata, Hiroaki Yano ( U. of Tsukuba )
      Artificial Life in Haptic Virtual Environment

3  Juli Yamashita, Yukio Fukui ( AIST, MITI )
      A Direct Deformation Method of Free Forms for CAD Interface

3  Masahiro Ishii, Makoto Sato ( Tokyo Inst. of Technology ) A 3D
      Interface Device with Force Feedback for Pick-and-Place Tasks

3  Martin Buss, Hideki Hashimoto ( U. of Tokyo )
      Intelligent Cooperative Manipulation Using Dynamic Force Simulator

O  Matthew Regan, Ronald Pose ( Monash U. )
      An Architecture for Orientation Mapping Post Rendering

O  Eliot Handelman ( Princeton U. )
      Permeable Space - a Language of Virtual Perception

I find the Proceedings to be an excellent and interesting source of
information. It is not a real book; it is a set of photocopies of
A4-size papers, which have been heat-bound. ( Note: that SIAM Japan and
other Japanese organizations use the same type of binding. )

________________________________________________________________________


3. ICAT '93 Video ( review and comparison with
                    the video:  Virtual Reality '93 (JTTAS) )

In addition to the proceedings, the conference package contained a video
of some of the VR, Tele-existence and artificial life systems described
in the papers. Some of the topics overlap with those which appear in the
video: VIRTUAL REALITY '93  (VR '93) distributed by the Japan Technology
Transfer Association. Although the clips from the two videos emphasize
different aspects in the same laboratories, they are similar enough to
make one regret the purchase of both. The contents of both are compared
and outlined below.

ICAT '93 Video:

The ICAT '93 video is, as advertised, a video supplement to the usual
printed proceedings. It shows the movements and inner workings of some
of the tools described in the papers, and is a valuable plus to reader.
Unfortunately, the quality of the clips in the ICAT'93 video vary quite
a bit. At the risk of sounding overly critical and somewhat
disrespectful, I offer the following notes of caution ( for those who
may be contemplating a purchase ) :
  - Clips 1, 2 and 5 are too long and could use some streamlining
    and/or editing.
  - Some clips suffer from poor audio recording.
  - All of the narrations may be a little difficult to understand
    by those who are not used to listening to non-native speakers.
  - Although the music in the fourth clip helps to pick up the pace,
    it becomes very distracting when combined with an Indian accent.

Contents: ICAT'93, Video Proceedings ( ~18 minutes )

1. Susumu Tachi, Kenichi Yasuda ( RCAST, U. of Tokyo )
   Evaluation Experiments of Tele-Existence Manipulation System
   ( narration by Japanese male )

   Tele-existence is defined and a method for quantitative comparison
   of tele-existence systems is proposed. The ability to complete tasks
   under different operating conditions is studied. Results are listed
   below in order of best to worst.
   - Binocular ( Stereo ) Vision + Head Mount Display (HMD)
   - Non-stereo ( 1-Camera ) Vision + HMD
   - Robot Mounted Camera, no HMD
   - Outside ( Remote ) Camera, no HMD
   The ability of Robots to handle complex situtations, e.g. moving
   fragile objects ( eggs ) is shown.
   ( Note: Same system appears in VR '93 video, similar egg scenario )

2. Hiroo Iwata, Hiroaki Yano
   ( Inst. for Enginering Mechanics, U. of Tsukuba )
   Artificial Life in Haptic Virtual Environment
   ( no narration )

   Artificial life is illustrated through the example: plant growth.
   Contents: Examples of growth and reversal of plants
             Examples of modified ramification frequency branching
             Chopping off branches by user

3. Juli Yamashita, Yukio Fukui
   ( Nat'l Inst. Bioscience and Human Technology, AIST, MITI )
   A Direct Deformation Method of Free Forms for CAD Interface
   ( narration by Japanese female )

   The clip shows how DDM methods and tools may be used by direct
   manipulation, freeing the user from mathematical details.
   Close-up/zoom and reduction features are also demonstrated.
   Future work on DDM + 3D input devices is discussed using
   a 3-D wire-mesh graph.
   ( Note: Same system appears in VR '93 video )

4. Masahiro Ishii, Makoto Sato
   ( Precision  and Intelligence Lab., Tokyo Inst. of Technology )
   A 3D Interface Device with Force Feedback for Pick-and-Place Tasks
   ( narration by Indian male )

   The SPIDAR systems are shown.
   ( Note: Same system appears in VR '93 video )

5. Martin Buss, Hideki Hashimoto
   ( Inst. of Industrial Science, U. of Tokyo )
   Intelligent Cooperative Manipulation Using Dynamic Force Simulator
   ( narration by Japanese female )

   Shows how robot hand is trained in two phases:
   1. Skill acquisition phase: a human manipulates a hand with sensors
      and the movements are recorded.
   2. Skill transfer phase: the movements recorded in phase 1 are
      transferred to a robotic hand. The results appear on a CG screen.

Virtual Reality (VR) '93 Video:

VR '93 consists of short interviews with professors managing the
laboratory followed by narrations describing various tools in the labs.
It is more of a tutorial and/or a show ( e.g. a Carl Sagan drama or NOVA
episode ) than a scientific research document. There is no hard-copy
documentation or even a pamphlet to accompany the tape.  Again, at the
risk of sounding overly critical and/or disrespectful, I noted that that
most scientifically inclined viewers seemed to find the first few
chapters of VR '93 to be excessively long, drawn-out and uninspiring,
and many felt that the interesting material could have been covered in
under 10 minutes, something to keep in mind when considering the
purchase of this video. VR'93 is distributed by JTTAS and is expensive:
10,000 yen ( covers postage and delivery for domestic orders, i.e.
within Japan. ) Details about costs for postage and handling for
overseas orders may be discussed on a case-by-case basis via surface
mail or FAX with Mr. Hara at:

Japan Technology Transfer Association (JTTAS)
2-11-2 Nagata-cho, Chiyoda-ku, Tokyo 100 Japan
tel: 03-3597-8220, fax: 3597-8224

Please note that the JTTAS recently moved to a new location so that the
address and numbers are different from those posted in a recent
information bulletin by Dr. Kahaner ["vr-video.93", 25 May 1993]. Phone
inquiries should be made in Japanese. If your inquiries elicit no
response, please note that my experience: it took me 4 phone call and 3
fax promises over a 6 week period to obtain a copy of VR '93. I then
received a double billing.

Contents: Virtual Reality '93, Vol. 1 ( 30 minutes )

Introduction (00:00-00.30)

Chapter 1 (00.30-06:43) Prof. Michitaka Hirose,
   Dept. of Mechano-Informatics, University of Tokyo

   Contents: introduction to VR devices, e.g.
   virtual dome, see-through HMDs, force feedback,
   air flow display, handling objects in virtual space

Chapter 2 (06:44-11.12) Dr. Makoto Sato,
   Precision Machinery and Electronics, Tokyo Inst. of Technology

   SPIDAR, SPIDAR III, POWER SPIDAR systems are shown.
   ( Note: Same sytem appears in ICAT '93 video )

Chapter 3 (11.13-16:37) Dr. Hiroo Iwata
   Inst. of Engineering Mechanics, Tsukuba University

   A desk-top and desk-side force display, a texture display, and a
   walk-through simulator are shown. It is puzzling that the walk-
   through system actually requires a person to walk on an exercise-type
   machine when a simpler device, e.g. a mouse could be used for input.
   ( Note: Hiroo contributed Haptic Life clip to ICAT '93 video )

Chapter 4 (16:38-21:04) Dr. Yukio Fukui
   Info. Factors Lab, Human Environment Systems Dept., AIST, MITI

   A force-feedback device using an XY recorder, a system for direct
   manipulation of form & curved lines and surfaces, a multi-modal
   integrative mouse, and sensor glove are shown.

Chapter 5 (21.05-30:00) RCAST, University of Tokyo
   Contents: a clip on a tele-existence mater-slave system
             a clip on tele-existence in a virtual environment
   ( Note: Same system as in ICAT '93 video, same example with egg. )

4. Impressions and Tidbits on the Conference.

True to its title, the the two topics which dominated the conference
were: 1. Virtual Reality and 2. Tele-existence. Most of the material
could be classified as either software or hardware ( robot, input or
display ). The Tele-existence category pretty much equalled Tachi's
work. The video clips are indispensible for fully appreciating the
research at RCAST. Regarding the hardware: the input device category was
dominated by the Japanese with their haptic, camera and wand devices.
Again, they are best described through their video clips rather than by
text. Unfortunately, one which does not appear is the wand, a magic
device which allows the user to create a world with a flick of a switch.
A poetic researcher commented, "wand-do de pick-ku shite, VR no sekai ni
ire-te-iku". In the display category, three head-mounts were presented:
EyeGen3, SONY Vizer Tron, and Olympus. All claimed to have 370,000
pixels and a 30-degree view range. I only received feedback on the
EyeGen3 and SONY devices. The 30-degree angle seems to be too small to
give the user a feeling of being immersed in a different world. The
EyeGen3 system had a poor screen image and was slow in reacting to move-
ments. SONY VT had a better screen image and faster response time.

CMU's Distributed VR presentation was appreciated by the Japanese as a
realistic and inexpensive idea. ATR presented an interesting Virtual
Space Teleconferencing System ( all Japanese Denki-makers seem to be
developing or trying to develop one ).

Ellis's investigations on the head movements ( 3-degrees of freedom with
a 60-degree range ) inspired some interesting informal discussions after
the conference. Will the results of the study only be relevant to
Americans ? Are there body-size, cultural or racial differences in
movements ? Harashima's work on facial expressions was also well
received, even by some skeptic-types.

Although the hardware experts revealed a number of interesting tools and
gadgets, software undoubtedly stole the show. Michael Zyda's real-time
graphics game dazzled attendees, including our local graphics experts at
IBM. For those lucky graphics and war game buffs who plan to attend this
summer's SIGGRAPH, rumour has it that Zyda will have a demo booth in
sunny Anaheim. The game lives in a set of high end Silicon Graphics
Systems and allows up to 500 players to shoot at each other. Here in
Japan, where unusually high proportions of children and adults love
amusement parks, pachinko parlors and video games, it was not surprising
that a heated discussion on business opportunities with a modified
version of Zyda's system began during a coffee break. One clever
Japanese corporate manager noted that the profits from a DOD start-up
venture might be used to eliminate the U.S. deficit. Discussion about
the game also inspired scientists to dream up new approaches to many new
problems for connecting 10,000 players ( or more ! ) in one system. For
instance, the development of an efficient message passing mechanism and
space alone poses challenging new questions. Seamless, or real-time
display without delay or hesitation poses another challenge.

Panel Discussion: ( with Ellis, Hirose, Tachi, Zyda, Iwata )

Listed below are some tidbits from attendees.

Ellis commented that VR is not a completely new subject; studies on the
relationships between man and machine have been developing over many
years. Costs and benefits of VR technologies will undoubtedly influence
and will be influenced by potential applications.

Hirose commented that the introduction of powerful and inexpensive
systems, such as the $5,000 IRIS system to be introduced mext month in
Japan, will open new doorways of imagination. Young and small users will
likely conjure up breathtakingly new and innovative applications.

Zyda's game as well as administrative abilities impressed one attendee
who noted that Zyda serves on the National Academy of Sciences commitee
on Virtual Reality. His key words: Hypermedia, Info. Node and Warp also
seem to have caught people's imagination.

Perhaps the Japanese are all too familiar with Iwata and Tachi to have
been left with a powerful new impression after the panel discussion.  An
interview with a foreign attendee would undoubtedly have given an
interesting and different view of the same panel discussion.