How Do Magic Eye Pictures Work?

Magic Eye's granddaddy was the random dot stereogram invented by neuroscientist and psychologist Bela Julesz in 1959 to test people’s ability to see in 3D. Julesz would generate one image of uniform, randomly distributed dots. Then, he’d select a circular area of dots within the image and shift that area slightly in a second image. Someone viewing the two pictures side by side perceive a circle floating above the background, even though the random dots had no depth cues. This supported his idea that depth perception happened in the brain, and not in the eye itself.

Twenty years later, a student of Julesz’s named Chris Tyler and computer programmer Maureen Clarke, discovered that the same thing could be done with just a single image.

Their research revealed what was happening in the eyes and brain when viewers looked at stereograms. When presented with an image like this, your eyes might each look at two different points, but because the image is a repeating pattern, the brain is tricked into thinking that the two spots are the same thing. The brain then perceives depth, with the two points as being on a virtual plane behind the pattern.

Magic Eye – which got its start in 1991 when engineer Tom Baccei, 3D artist Cheri Smith and programmer Bob Salitsky began building on Julesz’s and Tyler’s research – works by manipulating a repeating pattern to control the perceived depth and hide a three dimensional image in a two dimensional pattern.

A Magic Eye image starts with a programmer creating the hidden image (a schooner, for example) as a grayscale, smooth gradient depth map where dark points that should be furthest away are darker and closer points are in lighter shades. Then, they create a 2D pattern to camouflage that image. Finally, a computer program using a Magic Eye-patented algorithm takes the image model and the pattern and orients repeating patterns to the intended depth of the hidden image. When someone looks at a Magic Eye, the repeating pattern feeds the brain the depth information encoded into it, and the brain perceives the hidden picture.

More from mental_floss...

January 5, 2012 - 7:30am
webby
submit to reddit