Glasses-free 3D displays

Engineers have come up with a way to deliver 3D images on mobiles, televisions and tablet computer screens without the need for tinted specs
20 March 2013


Engineers at Hewlett-Packard have come up with a way to deliver 3D images on mobiles, televisions and tablet computer screens without the need 3D gogglesfor users to don a daft pair of tinted glasses.

We see the world in three dimensions because we have two eyes that are separated by about 6cm. Each eye sees, and feeds to the brain, a slightly different but overlapping view of the world. The brain stitches these two inputs together to create the three dimensional experience we perceive.

To recreate this effect using a flat screen, two separate images need to be fed to each eye individually.

Traditionally, this has been done using polarised light and a set of glasses that selectively filter the images destined for each eye.

An alternative approach sends two images from a screen but needs the users head to be in a very limited set of positions in order to appreciate the 3D image. The disadvantages of such systems are obvious.

But HP Laboratories scientist David Fattal and his team have developed a clever way around the problem, creating what's known as an autostereoscopic display.

Their solution is to use microscopic etching techniques to produce a matrix of tiny gratings that can sit atop the display pixels and control precisely the direction taken by light rays leaving the screen surface. A computer is used to turn the pixels on an off in the right order to create the 3D illusion.

In this way a much larger number of viewing zones - presently 14 but 64 should be feasible - can be created, giving users the perception of smooth stereoscopic images.

Using this technology, tilting a mobile phone displaying an image of someone's face, for example, would allow an observer to see around the back of the subject's head.

The manufacturing technique is actually quite simple, and David Fattal estimates that it will be available in the marketplace within 2 years.

More problematic, though, will be pursuading video producers to shoot their footage using the 14, or even 64, different cameras that will be needed to supply the multiple iamge sources the system requires to render the images!

As Cambridge University computer scientist Neil Dodgson puts it in an accompanying commentary in Nature, where the research is published, "All that remains is the more nebulous question of whether humans want or need 3D displays..."


Add a comment