Second LCD could help with VR’s nausea problem

According to researchers at Stanford University, one of reasons people get motion sickness while wearing virtual reality headsets is that in real life, our eyes get more light bouncing back from near-by objects than from distant ones. In a typical virtual reality headset, however, all the light rays are coming from the same screen, at the same intensity.

As a result, virtual reality views seem flat and fake — and can help contribute to the nausea that some people experience.

According to Stanford professor Gordon Wetzstein, there’s a solution to the problem, and it’s not too complicated. His team at the Stanford Computational Imaging Group has created a headset with a second, transparent, liquid crystal display screen in front of the regular one.

The second screen, in combination with a processing unit, allows for different amount of light to reach the eye, based on the virtual distance of the objects in the scene.

“Our insight is that only two LCD’s are required, thus providing an inexpensive solution,” said Fu-Chung Huang, who is also a visiting scientist collaborating with Wetzstein at Stanford. Huang also works at graphics technology company Nvidia Corp. as a researcher. “We enable the user to really focus in virtual space.”

The extra screen costs about $35, and the driver board another $30, but bulk purchasing — and the continuing fall in the price of all technology — will probably drive the cost down quickly.

The headset was built from scratch, using the Adafruit 3D printable virtual reality headset case, and lenses and LCDs bought on eBay.

(Image courtesy Stanford Computational Imaging Group.)

(Image courtesy Stanford Computational Imaging Group.)

And this “is really just the tip of the iceberg,” Wetzstein told Hypergrid Business.

The team will be presenting their research at the SIGGRAPH 2015 in Los Angeles next month, and the paper is also available online here.

The research is still in its early stages, however. It could take a while for the technology to be commercialized, and higher resolution will entail increased costs, said Huang, and fundamental challenges like diffraction that need to be overcome.

“People will going to explore further along this direction eventually, because really it gives what our eyes expect to see naturally,” he said. “People are going to like it once they experience the technology.”

 

 

 

Related Posts

Matt Kaiser

Matt Kaiser is a freelance technology journalist based in Japan. See his profile here.

  • This may be true, and it may work. But I must say that I am a bit sceptical, because there are a lot of situations where the background is illuminated, and the foreground is dark, and this makes nobody sick.

    Until now, the main source of sickness is the discrepancy between the movements that the eyes see, and the movements the inner ear perceives. And either way this discrepancy happens, we are sick: seeing movements while our body is not moving (as in virtual reality), or feeling movement while we don’t see them (as in a boat cabin). We can even be sick with a classical screen, if we do a succession of sharp turns. In the beginning I was quickly sick like this, but I get accustomed.

    Solutions to this are not simple, but they revolve around a better immersion, where our physical body must actually perform movements, before our virtual body copies these movements. This way what the eyes see matches what the ear feels.
    Simple to say, but more complicated than it looks: internet delays will create a time lag between the two experiences. It is not clear yet how much of such lag we can tolerate before being sick (that would be simple to study, though) but the solutions are clearly in the local loop, rather than in the Internet loop.
    Still, we need to have fast actuation of such informations as the position of the head. Since most today headsets work with a cellphone, they use the inertial unit of the cellphone to compute the position of the camera in the (local representation of the) scene. But cellphones are not designed for this, and very likely the sample rate of their inertial unit is too low. So that clearly a headset must have its own inertial unit integrated circuit, instead of relying on an external device.

  • LaeMing Ai

    I am wondering why you can’t just change the render engine to increase the brightness of foreground objects relative to the background?

  • Anna Lorentzson (cyberserenity

    Probably the biggest problem with VR is your own brain. I have found that as you use the headset more and more your brain learn how to. And the motion sickness goes away. But it is nice that there are people who continue the research. I think that getting VR out to the masses will be a challenge.

  • Mircea Kitsune

    I don’t have a VR headset yet, but I did notice this problem with red-cyan glasses: Although you can focus your eyes on various parts of a stereoscopic image, objects which are out of focus do not become blurry like they would in real life. Even if you can put each separate image on one eye, that can only achieve half of the stereoscopy effect!

    The idea to use a second screen to change the light intensity sounds great and pretty ingenious. I do however worry whether its functionality can be implemented generically, so the same code works for all VR headsets. Since Oculus decided to hugely delay the release of the Rift to next year (instead of this summer), they’ll hopefully have time to integrate this fix and make the delay worth it.

  • I suspected that the statement “eye receives more light from the background” is not the correct explanation on how this device works. Probably Mircea (next post) is right: the second LCD introduces a depth of field, forcing the eye to accomodate on close or distant objects as it does in the physical world. Weither or not this alleviates 3D sickness is to be tested. With my opinion messing with accomodation rather produces pain in the eye, blury vision, etc.

    Anyway, even if we get accustomed with time to motion sickness, it is important to find fixes which work for beginners too. Extensive studies are needed, and probably the solution will not be obvious, or it will entail a significant increase in price, such as having an inertial unit integrated in the head set (see my post under)