MAGIC EYE 

Discover how Magic Eye images work by learning how to create your own pictures with hidden 3D objects. Mike Bedford’s behind the visual wizardry.

Credit: www.easystereogrambuilder.com

OUR EXPERT

Mike Bedford has an enduring fascination with 3D photography, but, he admits, this is the first time that he’s understood how Magic Eye images work.

A t first sight a Magic Eye picture, or an autostereogram to give it its generic moniker, looks like a patterned image that repeats several times across its width. However, if you stare at them for a while, a three-dimensional object will eventually appear to pop out of the page.

Here we’re going to see how to generate your own Magic Eye images. But while it’s interesting to produce these astonishing pictures using readily available software – and we’ll show you how to do that – we’re going to start with a few rather more manual exercises. By delving into the underlying principles in this way, you’ll gain an inkling of the magic that fools our eyes.

Perceiving depth

When talking about 3D images we’re thinking of images that are able to reproduce one particular visual characteristic, namely stereoscopy. Putting that in plain English, this refers to us seeing a scene from two slightly different viewpoints – one from each of our eyes – in such a way that our brains merge those two images and, in so doing, provide a much stronger depth perception than is present in 2D images. Returning to Magic Eye images, it’s stereoscopy that makes their magic possible.

Generally, we don’t notice the two slightly different images that our eyes capture, because our brains do all the processing subconsciously. However, to get a better feel for this important visual effect, here’s a simple exercise you can try out. Hold the index finger of one hand at arm’s length and hold the index finger of your other hand much closer to your eyes. Now, try looking at this scene with just one eye at a time, and you’ll see how the horizontal position between the two fingers shifts as you swap between your eyes.

Taking this one stage further, we’re going to delve into stereo pairs or, in other words, pairs of images.These could be either photographs or computergenerated images that represent what each eye would see of a scene.

Having done the “fingers in the air” exercise, if you look at the stereo pair of the canal lock (below) you’ll be able to appreciate how the two images differ, even though the horizontal shift isn’t nearly as great.

In photography, the two images can be generated by taking two photos, moving the camera by a small horizontal distance between the two shots. Similarly, 3D modelling software can capture 2D images for two horizontally different viewpoints. Stereo pairs are viewed in such a way that our left eye sees only the left-eye image and the right eye sees only the right eye image.

Instruments called stereoscopes can achieve this using simple optics. However, because we’re going to be using a free-viewing technique later in this article – one that doesn’t require an optical aid – this is a good time to start. So, using the stereo pair, try out the viewing technique (see box, facing page). Once you’ve learned the technique, you’ll be able to see a fully three dimensional image of the scene represented by the stereo pair.