Special Report: 3D Video in Pro AV
In part one of a two-part special report, our video expert details how 3D video works and which technologies work best depending on need.
3D IN THE REAL WORLD
3Dor more accurately, stereoscopic imaging—comes in different forms (anaglyph, passive, active, autostereo) and can be used to show everything from black-and-white photographs to feature films and TV programs. The technology attempts to capture the depth cues that our eyes use in the real world and reproduce them on flat video screens.
When we look at objects around us, our left and right eyes see two slightly different views. As we fix on a specific object, both eyes focus on it. Then our brain combines the two offset views into one image (referred to as convergence) in order to process depth information. This two-step process is known as stereopsis and allows us to determine the relative shape, height, width, depth, and distance of objects within our field of view.
We use other visual cues to round out our stereoscopic vision, including shading (how light falls on an object), textural gradients, interposition, and perspective. Yet another more complex set of cues comes from motion parallax, or how the relative positions of one object to another change as we move past and around them.
The process of interpreting these visual cues is largely intuitive. We don’t have to think long about what we see to determine the size and distance of objects in our field of view. It stands to reason that a stereoscopic display should emulate the visual cues we see every day. Sounds simple, but it isn’t. The reason is that with a stereoscopic display, the focal point is always the screen surfacenot anywhere in front of or behind it. That restricts the degree to which we can accomplish 3D effects. In a person with normal binocular vision, depth cues are created by the difference between what our left and right eyes see, a difference known as retinal disparity, and the ability of our brain to focus on (accommodate) and blend those images into one (convergence).
The difference between what our left and right eyes see is small (unless we’ve got our face right up to the screen). Our mind takes the images captured by both eyes and overlays them in registration, creating a single image. Our eyes focus and converge at the same point, a condition known as zero parallax. In order to create stereoscopic images, we must induce some degree of image parallax to simulate retinal disparity, and then correct it with the appropriate eyewear, resulting in a 3D image.
The degree of parallax is not constant. It changes as objects in a 3D scene move closer or farther away, just as retinal disparity would change in the real world. The best 3D content uses this effect sparingly and mimics the convergence and accommodation processes to minimize eyestrain. The worst 3D content pushes parallax to the extreme, and while that results in some intense in-your-face effects, it also causes headaches, vertigo, and in some cases, nausea.