This story is over 5 years old.


Plasma-Based Midair Displays Are a Beautiful Techno-Future

Even if they are still kind of dangerous.
Image: Ochiai et al

That's sort of the future-UI dream, isn't it? Displays that are just hanging there, unmoored to circuitry and glass and batteries—all of the scrolling and selecting one could desire at the flick of a mid-air finger. A vaporous interface that would be just as well suited to Minority Report's precrime division as it would be to Candy Crush Saga.

Holography seems like a natural way to accomplish this, but researchers have been working hard on an alternative of sorts that doesn't rely on visual tricks or illusions and instead suspends pixels literally in space via laser-induced plasma displays. The technology, which has been around for several years in more primitive not-terribly-safe or interactive forms, is based on the ionization of air (or water) particles via nanosecond or femtosecond laser beams. Now, researchers at the University of Tsukuba's Digital Nature Group and the University of Tokyo have added some crucial refinements to the concept, allowing for the first time midair plasma displays that can be safely touched. Their work is described in a paper posted last week to the arXiv preprint server and set to be published in the the ACM Transactions on Graphics.


The group's creation is more properly known as a volumetric display, which is usually taken to mean a 3D image that's created without the aid of screens or visual effects through the direct manipulation of some sort of medium, like air or water. Crucially, a volumetric display can be viewed from any angle, unlike a traditional hologram. Here, the end result are "fairy lights" as seen in the video below. (Note that study co-author Yoichi Ochiai is a new-media artist as well as a computer scientist.) The fairy lights are rendered in "voxels," which are like pixels but with the addition of an extra dimension and, thus, volume: "volume" plus "picture" plus "element." A CT scan is rendered in voxels, for example.

The catch with rendering plasmonic voxels in mid-air is that it takes a high-intensity beam to do it. The basic idea is that some small patch of air molecules are being cooked until they start emitting light and if you were to, say, stick a finger into that voxel, you might find that the finger winds up getting cooked instead of the intended medium.

Ochiai and group seem to have found a happy (or happier) medium in ultrashort femtosecond laser pulses. These are pulses on the order of quadrillionths of a second. (Femtosecond lasers are used in applications ranging from medicine to heavy industry as an "ablation" tool, e.g. an ultraprecise way to slice stuff up.) The safety of their system was assessed using leather as a stand-in for human skin.


"Plasma has high energy and can be harmful to humans," the current paper explains. "However a femtosecond pulse is an ultrashort pulse laser, which is used for non-heat breaking for industrial purposes. It is also used for ultra-short scale fabrication of sub-micrometer order. Thus, we supposed that such pulses may not damage human skin seriously. In addition, our display scans a 3D space very rapidly, therefore, the laser spot does not remain at a specific point for a long period."

What the group found is that for durations of less than 2,000 milliseconds (or 2,000 pulses), there was no heat damage incurred on the leather but the pulses left supertiny holes (below) on the order of micrometers in diameter. They conclude that the system is thus at least "somewhat" safe and certainly safer than the more usual nanometer laser alternatives.

There are other potential problems with the technology besides, uh, burning holes through skin. For one thing, the generation of plasma with lasers is a nonlinear phenomenon. Essentially, this means that as more and more energy is added to the system, the response might grow exponentially or unpredictably. The system, in other words, may be unstable.

Image: Ochiai et al

The overall pitch is still cool, even if the experiments really only amount to a kind of/sort of proof-of-concept. "If laser-induced plasma aerial images were made available, many useful applications such as augmented reality (AR), aerial user interfaces, volumetric images could be produced," Ochiai and co. write. "This would be a highly effective display for the expression of three-dimensional information. Volumetric expression has considerable merit because the content scale corresponds to the human body; therefore, this technology could be usefully applied to wearable materials and spatial user interactions."