A Single Vision: Closing One Eye to VR

By Will Freeman,


"VR is all about one eye," asserts Unity's global director of evangelism, Carl Callewaert.

It's a contentious claim. After all, virtual reality as a means to fully explore 3D realms is arguably defined by serving a pair of user eyes. Even the most accessible written introduction to VR development is likely to pass a basic mention of implementing left-eye and right-eye images. Because VR – surely – is all about two eyes?

"You might not agree with me," Callewaert concedes. "You might think it's about two-eyes, but I want to take you on a journey."

That journey starts with the fundamentals. Our eyes, Callewaert explains, don't knowingly differentiate the real and the virtual. Our eyes, after all, are just input sensors for the brain, telling it what they see. It is our grey matter that we as humans use to make distinctions about the real and the virtual.

"We still understand space and depth with one eye," Callewaert continues. "If I close one eye, I can still see what is in front of me."

That, he says, comes from practice.

"As a baby you immediately know to reach and grab for objects. And still you are learning to see. It takes somebody easily many years to fully understand depth."

The human eye uses myriad clues to understand scale and depth, even when there appears to be little to go on, as seen here in the 1971 movie Vanishing Point

Certainly, closing one eye does not make the real world around you suddenly appear to be entirely two-dimensional. And we are, Callewaert suggests, entirely capable of seeing through the trick when a holidaymaker poses far in front of the Leaning Tower of Pisa, hand outstretched as if they are holding the structure in place. It's a bad joke about scale we all know, whether using one eye or two, but what this has to do with VR content development is less obvious.


There are, it turns out, a good number of 'monocular cues' that we use to understand depth and scale in the world around us – real or virtual – and we only need one eye to understand them. The apparent size of a familiar object in our field of view, for example, helps a great deal, as does the familiar illusion of perspective and vanishing point. We might use cars at spaces along a road stretching away from us to gain a sense of scale of the objects around them.

There are less obvious examples of monocular cues too, such as aerial perspective, where light scattering caused by the atmosphere means objects that are further away demonstrate lower luminance contrast, as well as lower color saturation. We may not know about every monocular cue, but we use them every day, as long as we open even one eye.

A close-up view of the Eiffel Tower demonstrates several techniques relevant to VR

Perhaps surprisingly, there are far less binocular – two-eye – cues. Stereopsis uses the fact our eyes are in different places to get a better sense of an object viewed from two points. Convergence, meanwhile, lets our brain make judgments about depth based on the distinct and separate movement of our two eyes looking at the same object. And then there's shadow stereopsis, which lets our brain harness shadows to translate the space around us. So as far as binocular vision, that is it.

Finally, we have the simple ability to move, craning out necks, for example, to get a better understanding of what we are seeing relates. If we could visit the moment a Leaning Tower of Pisa picture was taken, movement would let us quickly verify that the jovial tourist wasn't really resting a palm on the building's wall.


And that is the main leg of Callewaert's journey. These are tricks architects have used for centuries, and he believes VR developers should consider using them too, embracing monocular and movement cues to speak to the player about depth and space. It's a way to make things clearer in VR, and even allow for increased comfort, Callewaert offers.

"That's why, in VR, you should try and place objects in the background and the foreground, so users can move and understand what they are seeing," he says.

"And ask yourself; in VR, should you control the player by making them move so they can understand the space, or do you leave them standing still?"

That is an important reason you should allow you VR user movement, he adds.

"Ask somebody who visits St Peter's Basilica what they did, and most of them will say 'we walked around'. People move around to understand the space."

Visit Rome's Pantheon, or the ancient Greek Parthenon, or any number of architectural icons, and you will see falsified monocular cues; tricks of shadow, scale, perspective and light. You might even notice the Eifel Tower's paint hue changes with the structure's height, to mark it out from the sky and emphasize presence. All these things Callewaert suggests worthy of study if you are approaching design for VR.

A closer look at the Parthenon reveals many useful tricks for VR devs

"These are beautiful buildings that give us spirit, and if you look at those buildings, they use one eye clues to do that," he says, later adding "one eye information will feed two eyes, and let you understand depth."

And he has a final piece of advice for VR developers, that captures the very point of considering VR as a pursuit that should consider the single eye.

"When you make a VR game, close one eye, test the game and play it," Callewaert insists. "If you don't have a sense of depth, look again at your depth cues."

Follow Carl Callewaert on Twitter: @CarlUnity


Lead image credit for this piece: Jean-Christophe BENOIST

Got it

We use cookies to ensure that we give you the best experience on our website.

Click here for more information.