Stereoscopic Display Technology Basics

A StereoMirror™ monitor consists of two AMLCD displays oriented at a fixed angle through use of a specially designed mounting stand. A passive beamsplitter mirror bisects the angle formed between the two monitors mounted on the stand.

The mirror has the following properties:

• 50% transmission
• 50% reflectance
• Non-birefringent
• Flat spectral response from 400 to 700 nm (visible spectrum)
• Abrasion-resistant hardcoat

An anti-reflective coating is deposited on the side opposite the transflective thin films to minimize secondary image reflection. There is a fine mechanical adjustment for the mirror angle between the two displays.

The objective of a stereoscopic display is to efficiently present a Left eye image to the Left eye that is isolated from a Right eye image presented to the Right eye. This allows the visual system to merge the two images resulting in the perception of depth, or stereopsis. In the StereoMirror™ design this stereo separation is achieved using the principle of conservation of polarization. Liquid crystal displays operate based on the ability of liquid crystal material to modulate plane-polarized light. The two AMLCD's in a StereoMirror monitor are manufactured so that the plane of polarization from the lower monitor (Left eye image) is perpendicular to the the polarization plane in the light path of the upper monitor (Right eye image). When stereo pair images from the two monitors are viewed through crossed-polarizing glasses, the user only sees the lower monitor image with the left eyepiece and the upper monitor image with the right eyepiece. The result is a single, fused stereoscopic image.

The left eye and right eye images are sent to their respective AMLCDs independently from an off-the-shelf dual-output graphics card. Presenting the stereo pair of images is accomplished using a software application that supports stereo viewing. This software is typically written to be compatible with either OpenGL or DirectX standards. It requires no additional modification for use with the StereoMirror™ monitor design. Any software application that uses the OpenGL quad-buffered stereo features is compatible with the StereoMirror™. Quad buffered stereo is a feature of the OpenGL 3D graphics library that allows an application to define two separate Right/Left eye viewpoints instead of the normal single monoscopic viewpoint. The two viewpoints are defined to give the correct parallax separation for the proper stereo effect. Once the two viewpoints have been defined the 3D scene is rendered identically for each of the two viewpoints. Many commercial 3D applications already have stereo viewing modes using the OpenGL stereo features.

Since the upper display in the monitor is seen in reflection there is a need to mirror the image on that data path. This is accomplished in a number of ways:

• Software processing in the computer CPU
• Firmware manipulation in the graphics card
• An auxiliary signal processing board in the data path to the upper monitor (PCI mirror card or DVI Inverter Box)
• Manipulation in the monitor controller board

All of these possibilities have been demonstrated. Manipulation at the graphics card is the preferred option to minimize cost and the possibility of adding a delay between the two video paths. Driving a StereoMirror™ monitor is identical to driving a pair of projection displays used to show stereoscopic images with crossed polarizers in the two separate light paths.

Related Content

• Comparing Dual Stacked LCD Stereoscopic Monitors with Planar StereoMirror™ (pdf)
• Planar's StereoMirror Technology versus Red Rover/True 3Di (pdf)
• Stereopsis: Two Eyes, One 3D Image
• Exploring 3-D display technology
• 20 Reasons Why Planar's StereoMirror Offers Superb Stereoscopic Viewing (pdf)