PITTSBURGH -- Virtual reality, the technology of the early '90s, used headsets that transported people to a computer-generated faux-world. Here in the late '90s, VR's half-cousin, the wearable computer, also wants to deliver a world to the user. But this time, it's the real world they're serving up, using virtual information to augment reality, not replace it.
"Augmented reality involves overlaying information -- graphics and annotations -- onto the environment," said researcher Li-Te Cheng in his presentation Tuesday at the second annual International Symposium on Wearable Computers. "A doctor," for example, "might overlay instructions on how a medic should care for an accident victim."
Augmented reality proponents like Cheng hope the future of wearable computing can assist people in any environment, and provide an ideal combination of the real and the virtual.
Fields where researchers see virtual information being added to real-world environments include telemedicine, architecture, construction, and devices for the disabled. While the ideas can certainly be applied to stationary computer set-ups, augmented reality's most natural place is in the mobile, wearable computer world.
But more enticing was a scenario presented in a symposium paper by Sony Computer Science Laboratory's Jun Rekimoto. He sees augmented reality -- when implemented in normal eyeglasses -- making a walk down the street much more enticing.
"When you go outside for lunch, you find several virtual messages are floating in front of restaurants," Rekimoto's scenario goes. "Some are commercials, but many are messages that were created and attached by previous visitors, giving information and ratings on the restaurant. You select a restaurant by looking at these messages. While you are at lunch, you remember that you will have visitors at noon. You select an icon representing your office in the eyeglass, and attach a voice memo to it..."
While future possiblities of augmented reality sound a bit more appealing, Cheng outlined more immediate research where a remote expert interacting with a field worker (a staple ingredient in wearable computing demos) "augmented" what the field worker was doing. Equipped with a video camera, the worker's heads-up display shows annotations and messages added to the scene by his remote counterpart.
A successful augmented system ties wearable computers to augmented reality techniques and technology that allows a remote user to manipulate the image on a shared display. Circling someone in a scene, for example, or attaching virtual "sticky notes" that label objects in a scene.
While these technologies have been used to collaborate on plain old PC spreadsheets, wearable researchers have had to adapt to a much more fluid environment. Camera and shared mouse pointer systems are just a few areas where specialized research has been conducted to accommodate wearable devices.
"Because it's a wearable computer, you want the worker to move," Cheng said. "But the problem then becomes that information is always leaving and entering the scene."
The Multimedia Communications Lab at the Memorial University of Newfoundland, home to Cheng's research, came up with an interesting solution. As the mobile worker surveys a room, his camera delivers the images back to the remote expert. But instead of straight live video playback, software at the other end used individual frames of video to build a still composite image. With frames from various angles, the result was a quickly assembled panoramic "mosaic."
The effect is similar to taking several pictures of a scene and taping the prints together to build a panorama.
"[This system] allows free movement but divides time into 'registerable' shots," Cheng explained, "which can be turned into the 2D panoramic composite." The image is constructed in real-time, but it works fast enough for the expert to rapidly view a bigger portion of the room scene.
"It's still a work in progress... it's not exactly a real-time system. Our ideal is 3 or [so] frames per second."
With this more commanding vista -- in the case of a doctor, perhaps of an entire victim's body rather than just an arm or head -- the expert can annotate the image for viewing by both workers via shared displays. Sticky notes and drawing tools let the expert add text labels to objects in a scene and mark specific areas.
On a panorama of person's body, "a doctor could draw a circle on the guy's head," Cheng said.
A videotaped demo of the system showed a worker surveying a printer station. After the panning camera mounted on the remote user provided a panorama, the desktop user circled the printer and showed him how to replace the paper tray.
Maybe not such a wondrous example of reality, but it is augmented.
