Escape From Flatland

Holografika's 3-D display is so good that if you move your head, you can see around a corner

5 min read

BUDAPEST—I am standing in front of a large television screen that shows the image of a red pickup truck. Glare from the truck’s windshield obscures the cabin, so I move my head and find myself peering inside; I move my head back, and the glare returns. I remind myself that this is, indeed, a flat display—not a diorama.

It is, rather, three-dimensional imagery in the full sense of the word, on parade in the offices of Holografika, a small Hungarian company named after the laser-mediated technique that earned Hungary’s Dennis Kapor the 1971 Nobel Prize in physics.

”Yes, the history in Hungary, it played a role” in getting government support for research on this 3-D gadget, says Tibor Balogh, chief executive of Holografika. ”Mainly, though, it is because of a lucky idea.” Rather than try and fool my brain into the illusion of depth with stereoscopic views, or show me a series of frames that varies as I move my head, Balogh creates a virtual image just the way a CT scanner does, when it takes myriads of X-rays coming from different angles and deduces the shape of the object.

He hits a button and on comes a movie of three interlocking gears, rotating along orthogonal axes—upward, sideways, and inward. Move aside, and now you are seeing it from a different angle. Part of a drive shaft has come into view.

Parallax, Pixel, and Pinholes

The Holografika system can be considered, metaphorically, as a source light, a back screen with pinholes, and a front screen, also with pinholes. Each ray of light entering your eye has been sent there through two pinholes, as it were, via a filtering method in which all the rays trace back not merely to the actual light source, but even farther, to a virtual object—an image being modeled in three dimensions. Move your head, and a different set of pinholes conveys rays that appear to come from the same image but along a different angle.

Each eye gets a slightly different image, and therefore your brain constructs an image with depth, just as it would for stereoscopic pictures, like those seen through a slide viewer. However, the brain also gets a second sense of depth, from parallax—the changes in aspect that correlate with movements of the head. That is why prolonged exposure to stereoscopy without parallax can confuse the brain, and even turn the stomach.

”The old stereoscopic movies were for the most part disorienting,” Balogh says. ”You will never watch a television that makes you sick.”

In the early 1950s, there was a brief fad for stereoscopic movies, created by projecting two images through different color filters. To see one image with the left eye and the other with the right, viewers had to wear glasses with different colored lenses. To minimize the nausea, a clever director would use stereoscopy sparingly, as Alfred Hitchcock did in Dial M for Murder (1954). Hitchcock really ladled the 3-D magic on in the climactic scene, in which Grace Kelly, strangled by an assailant, thrusts her hand out into the viewer’s face, clutches at scissors, and uses them to stab her assailant in the back.

”Those old movies were for the most part disorienting,” Balogh says. ”You will never watch a television that makes you sick.”

Surgeon’s Friend, Gamer’s Passion

He hits another button, and I see the delicately branching tree of blood vessels in someone’s brain with an outpouching clearly visible. I look at it from one side, then from the other—yep, it’s an aneurism, all right. A neurosurgeon, looking at the same image, might project onto it the surrounding skull to get an idea of just how to get to the defect while causing the least damage to the brain.

An even more exciting possibility occurs to me, poor lowbrow that I am: this could be the world’s best gaming interface! Balogh seems slightly disappointed in me. Yes, he says, with a sigh, we have considered that market as well.

He takes me into another room where three engineers are fooling with the innards of the company’s invention, available so far only in test versions. Two are assembling light-emitting diodes (LEDs), the sources at the back of the display. The third one runs a program that brings to life a hulking figure that swings a club, while taking a step in my direction, a maneuver the figure repeats again and again. It’s not much of a movie, I complain. Yes, says Balogh, but this is no prerecorded movie—it’s an on-the-fly animation, calculated from the sort of raw data a program would get from a move I might have made in an interactive game.

”First, we have the bones,” he says, and I see a hulking skeleton swinging a stick. ”Then we have the skinning,” he continues, as the program figures out the stretching effect on upper layers—first, the figure’s skin, then its clothing. The beauty of it, he says, is that game companies can make the transition to his 3-D method rather easily, using much of the data that they generate for their 2-D displays but ”do not now fully exploit,” as he puts it.

Use the 3-D Data You Already Have

Right now, such a game player would set me back some 30 000 (about US $ 39 000), so I guess this scientific advance will be limited to the neurosurgeons for the time being. Several hospitals have indeed been looking at the prototype machine, which Holografika brought out in 2005. It is accepting small orders even now.

Doctors, too, will find it easy to snap the Holografika display onto existing diagnostic scanners, based on magnetic resonance, positron-emission, or plain, old X-ray tomography. The data from those machines reflects three dimensions anyway, so it is just a question of rendering it properly. That way, a surgeon could superimpose such a diagram on a person’s body to plan how best to reach the target organ.

The same goes for engineers trying to visualize the wiring in a new jetliner. Here, too, the existing CAD/CAM systems already compute in three dimensions, so their data need only be poured in to display a fully rounded device—those interlocking gears, for instance.

It could also help broadcasters in televising sporting events. ”Even five cameras, pointed at a football field, can be used to generate 3-D, by using computers to interpolate the data,” Balogh says. ”It would then be as if we had used 100 cameras. We could reconstruct a [soccer] game from the point of view of any player—say, the goalkeeper—even if we had no camera there! Now it might take a supercomputer, but soon�.”

Holografika recently brought out a bigger, sharper, longer-lived model using LEDs rather than arc lamps. They have the advantage of being smaller, cooler, and longer lasting. The new display also expands the viewing angle by one-sixth, to a total of 70 degrees.

Today, Balogh says, his company could not even handle an order of 100 000 units. Yet he looks forward to the day when rising demand for 3-D and falling unit prices make such volume possible. To meet the challenge, Holografika will need to forge alliances with major manufacturers. Meanwhile, Balogh says, it is planning a second round of financing, sometime in the second half of 2007.

This article is for IEEE members only. Join IEEE to access our full archive.

Join the world’s largest professional organization devoted to engineering and applied sciences and get access to all of Spectrum’s articles, podcasts, and special reports. Learn more →

If you're already an IEEE member, please sign in to continue reading.

Membership includes:

  • Get unlimited access to IEEE Spectrum content
  • Follow your favorite topics to create a personalized feed of IEEE Spectrum content
  • Save Spectrum articles to read later
  • Network with other technology professionals
  • Establish a professional profile
  • Create a group to share and collaborate on projects
  • Discover IEEE events and activities
  • Join and participate in discussions