The display uses so-called integral photography to generate its 3D images. A computer divides up the image into pixels that are either printed on photographic film or shown on a flat screen display. When passed through an array of lenses the result is a 3D image with a depth of 5.7 m or more in front of the display and 3.5 m or more behind the display.
"The integral photography technique has none of the inherent eye-straining problems associated with continuous viewing of binocular stereoscopic displays," University of Tokyo scientist Hongen Liao told Optics.org. "It is an ideal way to display autostereoscopic images."
Up until now the success of this technique has been limited by the accuracy of the lens array, which here is composed of 6 mm diameter glass lenslets arranged in a 35 x 35 hexagonal layout. As Liao explains, even a small error in the lens' arrangement can cause image overlap and blurring, which limits the precision and depth of the 3D image."
"Most of the autostereoscopic displays only have an image depth of several centimetres," said Liao. "To the best of our knowledge, there has been no report about producing an image with a depth of several metres."
Liao and his colleagues have found a way of preparing images that allow them to be viewed at long distances. Firstly, a computer pixelates the image to match the 3D display's lens array and projects this image on to a screen. The researchers then use the 3D display's lens array to capture this view on photographic film, creating a reference image that self compensates for any lens distortion or misplacement.
"Our method enables a display device to present virtually deviation-free and distortion-free 3D images," said Liao. "We are now developing an animated long viewing distance integral videography device."
The letter 'A' stands out in midair approximately 1 m from the screen. (credit: University of Tokyo, Japan)
Author
James Tyrrell is reporter on Optics.org and Opto & Laser Europe magazine.
