Restricting field of view to reduce motion sickness in VR
The immersion that virtual reality provides is one of its best selling points – and biggest drawbacks. Motion sickness is a common side effect of VR, where the brain has trouble reconciling the movement the eyes are seeing with the lack of motion the body is feeling. It's a problem that many are working to address, including teams from Microsoft and Stanford University. Now researchers at Columbia University have developed a new approach that involves masking the player's view to minimize the symptoms.
Nausea in VR is caused by the brain receiving mixed signals from the senses: the eyes are detecting movement, such as walking or driving, in the virtual environment, but the vestibular system in the inner ear, which relays our sense of motion and balance, doesn't match up because the player is in the one spot.
The Columbia researchers focused on the player's field of view (FOV) and found that by adjusting that visual range on the fly, VR sickness could be drastically reduced. Generally, that means simply decreasing the FOV, so with less visual cues the brain doesn't register as wide a discrepancy between sensory input.
Unfortunately, reducing the FOV too much can pull a user out of the VR experience. The normal human field of view is 180 degrees, but most VR headset displays show about 100 degrees. Lessening that further can make the whole thing feel like you're looking through a pair of binoculars.
To find a middle ground, the researchers at Columbia University's Engineering School developed a software system of "dynamic FOV restrictors" to automatically adjust the FOV depending on the situation, based on the user's speed and angular velocity in the virtual world. When the player is in motion in the virtual environment, the system partially obscures each eye with a soft-edged, circular, virtual cutout, reducing the FOV to minimize nausea. When there's less action, such as when the player is standing still, the view increases again, to allow for better immersion. Ideally, it does all this without the user even noticing.
The researchers tested their system over two days on 30 participants divided into two groups. One group used the FOV restrictors while exploring a VR environment on the first day and went without the next day, and the other group did the opposite. In the tests with the restrictors in place, participants reported feeling more comfortable, and stayed in the virtual world for longer.
What's more, two different versions of the system were used, one made to be more subtle than the other. Most users didn't even notice the less obvious one, and those that did unanimously preferred the experience with it.
Buoyed by the results, the researchers plan to continue looking into ways the FOV restrictors can help users feel more comfortable in virtual reality. They're interested in exploring different shapes and textures for the cutouts, as well as automating adjustment through other parameters, such as the heart rate of the user, and optical flow, which is the apparent motion of objects in a scene resulting from the relative motion between the observer and the scene.
It's interesting to note that this system is almost the exact opposite of the related Microsoft Research project. In that case, the FOV was increased to 170 degrees, much closer to the natural human range, by fitting out an Oculus Rift DK2 headset with LEDs to mimic our peripheral vision. With a more natural field of view, the VR experience feels more natural and as a result, is both more immersive and less nauseating.
The Stanford research project centered on the conflict between depth and focus as the sickness-inducing element. Current VR headsets only simulate depth through stereoscopic lenses, but the Stanford team created a prototype with multiple layers of screens built in, allowing the eyes to focus more freely.
With different projects tackling the issue, hopefully later generations of VR equipment will be more comfortable for the queasy.
The results of the Columbia University research were published in the journal IEEE, and the approach is discussed in the video below.
Source: Columbia University