Email to a friend
New video-conferencing method
cheaper, more sophisticated, developers say
Mitchell, News Editor
photo to enlarge
by L. Brian Stauffer
| Klara Nahrstedt, professor of computer science, is
collaborating on a new video-conferencing method that
is more sophisticated and cheaper. Graduate student
dancer Renata Sheppard is in the background.
— If only Fred Astaire and Ginger Rogers were around today to
take a spin with new technology being developed and tested by a team
of computer scientists in Illinois and California.
If they were, they’d be dancing circles around each other –
only from a considerable distance. That’s the beauty of Tele-immersive
Environments for EVErybody, or TEEVE, a system that’s being
test-driven simultaneously across thousands of miles this spring in
the labs of Klara Nahrstedt, a computer
science professor at the University of Illinois at Urbana-Champaign,
and Ruzena Bajcsy, a professor of computer science at the University
of California at Berkeley.
In technical terms, TEEVE is a distributed multi-tier application that
captures images using 3-D camera clusters and distributes them over
Internet2 (the network reserved for research and corporate clients),
compressing and decompressing the 3-D video streams, rendering them
into immersive video and displaying them on one or multiple large screens.
In layman’s terms, think of TEEVE as a turbocharged version of
videoconferencing, but with some very fancy new bells and whistles.
Most notably, Nahrstedt said, TEEVE makes it possible for people to
view their counterparts at remote sites from all angles.
And an important feature that sets it apart from other tele-immersive
video-conferencing systems currently being developed or used elsewhere
is its potential for delivering high-quality images and communications
using relatively inexpensive technology and COTS – or commercial-off-the-shelf
products and equipment.
“TEEVE is a great technology because it allows for more cost-effective
cyberspace communication of people in their full body size,” Nahrstedt
“This system is especially suited for learning new activities,
training and meeting in cyberspace if a physical activity is to be performed,”
The researchers also believe the technology is ideally suited for a
variety of entertainment-related purposes.
“With TEEVE we want to allow distributed artists such as dancers
to train, design new choreography and experiment with different movements
in the cyberspace,” she said, noting that TEEVE’s relatively
low price tag would be of special interest to artists, who typically
struggle to produce their work with limited funding.
This spring, Nahrstedt, Bajcsy and their research teams have been testing
the technology with the aid of two performers: U. of I. dance student
Renata Sheppard and U. of C. dance professor Lisa Wymore. In each experimental
pas de deux, Sheppard stretched and spun about before semi-circular
clusters of 3-D cameras in Nahrstedt’s lab on the Urbana-Champaign
campus, while Wymore executed her moves in Berkeley in a similar environment.
To date, Nahrstedt has been pleased with the results, which she pronounced
“exciting and excellent.”
“Both dancers met in the cyberspace, danced together and also
synchronized when dancing,” she said.
Among other potential applications, Nahrstedt expects TEEVE will, in
the not-too-distant future, allow for the following scenarios to take
• After accidents, medical patients and physiotherapists meet in cyberspace,
where the physiotherapist demonstrates muscle-strengthening exercises.
• Students are able to learn new sports or movement activities, such as
tai chi, even when living in remote locations where no local teacher
• While communicating with an elderly parent, adult children living far
away can more accurately assess a parent’s physical condition.
Nahrstedt predicts that it will be at least five to six years before
TEEVE and other tele-immersive 3-D multi-camera collaborative environments
are routinely used in university or corporate settings.
“Videoconferencing equipment – 2-D, single view –
has been available for the last eight to 10 years, and only maybe in
the last three years has it become common to have a conference room
equipped with polycom conferencing equipment or access grid or net-meeting
on a more regular basis.”
In the meantime, Nahrstedt said, she and her colleagues will continue
to design new software systems, protocols and hardware capabilities
as common new platforms, such as multi-core processors and better cameras,
A number of interesting research issues remain ripe for exploration
as well, she said. They include figuring out how to simplify human-computer
interfaces to allow people to customize their content and displays in
order to more easily process multiple views and large amounts of information,
and real-time processing and communication that would bring 3-D tele-immersive
content closer to the quality of current television and radio quality.
Another big goal for the researchers is to focus on ways to automate
the technology and make it more user-friendly.
Ultimately, she said, “the environments should be set up with
a push of a button – this is absolutely impossible at this point.”
Research on TEEVE was presented earlier this year at the Multimedia Computing
and Networking conference in San Jose, Calif., and featured in the February
issue of ProAV magazine. At Illinois, Nahrstedt’s research team
consists of graduate students Jigar Doshi, Jin Liang, Wanmin Wu, Zhenyu
Yang and Bin Yu. Working with Bajcsy at Berkeley are Ross Diankov and
Samuel Morris Johnston.
The research was funded by a grant from the National Science Foundation.