3-D reality without the real

Conventional 3-D displays force our eyes to do two conflicting things at the same time, for instance, pointing our eyes to look at something that appears to be far away, but having to focus the lenses of our eyes up close. But this unnatural eye pose is tiring and gives people headaches. It's easier and more natural for our eyes to focus at the same place that we're pointing them. Brian Schowengerdt at the University of Washington's Human Interface Technology Laboratory has a new 3-D display, called True 3D, that matches what's most easy on the eyes.

With the True 3D scanned-light display, light from different objects seems to come from different distances in space. This is made possible by a tiny stretchable mirror made of a thin membrane, just 10 millimeters across, coated with aluminum. The deformable mirror stretches on command to change the focus of each pixel of light as the display projects different objects. Just one tiny mirror can control all the pixels in the display as it scans by changing the focus of that beam very quickly -in this case, twice as fast-as the display refreshes. Viewers can converge their eyes and focus their eyes at the same distance - just like when viewing real objects. Also as in real life, no screen is needed to see the objects - the display changes the light's intensity and color, so "a high-resolution full-color picture can be painted right onto the retina," Schowengerdt said.

Brian Schowengerdt explains more about how our eyes work, and how the usual 3-D displays -- and his new display – trick our eyes into seeing things in 3-D.

How do our eyes see things far away?

Our eyes are separated in space by a few inches, so when viewing the real world they have slightly different viewpoints of the same object. You can verify this yourself by holding a finger on your left hand at arm's length, and a finger on your right hand in front of it and close to your face. If you close one eye and then the other, you can see that the two fingers move to the left and right.

To avoid "seeing double" when looking at nearby objects we have to *converge* (rotate inwards) our eyes, and to look at distant objects we have to deconverge our eyes and point them both straight ahead. This is true both when looking at real objects and objects presented in stereo "3-D" displays.

However, the focus of the eye is different between viewing these displays and viewing real objects. Because in the real world the objects are really at different distances, the light reflected from them (eventually reaching our retinas) comes from different distances. To focus the light from real objects we have to place the focus of the lenses of our eyes at the same distance to which we're pointing (converging) our eyes. Because we have to do this all the time in the real world, nature took a shortcut and reflexively-linked the focus of the eyes to the convergence of the eyes. When we point our eyes to one distance, we automatically try to focus them to the same distance.

How do conventional 3-D displays work?

Conventional "3-D" displays aren't really completely 3-D (they're just stereoscopic). They are really just two 2-D displays (like LCD screens), and by showing slightly different viewpoints to the left and right eye with the 2-D displays, the brain experiences the illusion (mostly) of seeing things in 3-D.

Conventional 3-D displays force our eyes to do two conflicting things at the same time, for instance pointing our eyes to look at something that appears to be far away, but having to focus the lenses of our eyes up close. This is tiring and gives people headaches. (It's a bit like sea sickness, when our sense of balance tells us that we are moving, but our eyes tell us that we are staying still. This mismatched sensory information makes us get sick.) It's easier and more natural for our eyes to focus at the same place that we're pointing our eyes, and it helps us to distinguish the far from the near (as you focus on an object, and the lens of your eye curves to bring it into focus, things at the same distance look clear, but things farther or nearer look blurry.)

Unfortunately, because stereo "3-D" displays really just use two 2-D displays, the light that represents objects at different distances in the displayed scene all comes from the same point in space--the surface of the 2-D display (e.g., LCD). The stereo nature of these displays still makes it necessary to dynamically change the distance to which the eyes are pointing, but viewers have to try to keep their focus fixed at one distance. This forced decoupling of the reflexively-linked eye convergence and focus causes eye fatigue and headaches, and makes the displays less realistic.

How does your scanned light display imitate the way our eyes really see?

The True 3D scanned-light display makes it so that the light from different objects comes from the correct distances in space. The deformable mirror dynamically changes the focus of the light as different objects are projected with the display, so those objects are optically-placed at the correct distances. With the True 3D display, viewers can converge (point their eyes) and focus their eyes at the same distance--just like when viewing real objects.

What is a deformable mirror? How does it differ from a normal mirror?

The deformable mirror is a MOEMS (Micro-Opto-Electro-Mechanical-Systems) device. MOEMS (and their cousins MEMS) are miniature devices that can be "printed" directly onto chips in batch processes. The mirror we're using consists of a 10 mm diameter thin silicon nitride membrane that is stretched across a circular frame. The membrane is coated with aluminum to make it into a mirror. An electrode is mounted behind the membrane. As voltage is applied across the membrane and electrode, the center of the membrane is electrostatically pulled toward the electrode. With no voltage, the mirror is stretched flat, and light reflected from its surface remains at the same focus level. When a voltage is applied, the mirror is shaped into a concave parabolic mirror, which increases the focus level of the light it reflects. Using the mirror, the focus of the scanned light beam can be dynamically adjusted to place different pixels in the display optically near or far away (or anywhere in between). The mirror we are currently using was made by OKO technologies in the Netherlands.

What is a scanned-light display? How is the deformable mirror used in the display?

A scanned light display works a bit like a television set. In a TV, an electron beam is scanned back and forth in a raster pattern across the glass screen. In a scanned-light display, a beam of light is scanned in a raster pattern--but rather than scanning the beam across an intermediate screen, the beam is scanned directly onto the retinas of the viewer's eyes. By varying the light intensity and color of the beam as it scans onto different parts of the retina, a high-resolution full-color picture can be painted right onto the retina. The viewer perceives the image to be floating out in space in front of him/her, but really the image doesn't exist anywhere but in the eye of the viewer. The core technology of the scanned-light display was invented at our Human Interface Technology Lab at the University of Washington in 1996 (and was called the Virtual Retinal Display, or VRD), but the technology is also related to that of scanned-light opthalmoscopes (SLOs), which were invented in 1989 by Robert Webb to take pictures of retinas.

In the True 3D display that I'll be presenting [at the Frontiers in Optics conference, a meeting of the Optical Society of America], I've incorporated a deformable membrane mirror into the display, so that the focus of the beam of light can be dynamically varied as the beam scans different pixels onto the retina. In doing so, I can generate complex multi-focal images with different objects appearing to float at different locations in space.

More information

Martha J. Heil
mheil@aip.org
301-209-3088