NLM Home Page VHP Home Page


Next: Methods Up: Title Page Previous: Full Text Index Index: Full Text Index Contents: Conference Page 

Introduction

      The principles of stereoscopy are not new. In fact, they were known even before photography was invented. Charles Wheatstone produced some of the first known stereoscopic drawings of simple objects back in 1832. With the discovery of photography a few years later, stereo images quickly became a novelty of the Victorian era. Since that time, numerous techniques have been invented to create and display stereo images. Although these inventions differ greatly, they all use the same basic technique to produce the illusion of depth from pairs of two-dimensional (2D) images. Depth is produced primarily from the display of an object from two slightly different perspectives. These perspectives correspond to the views from the left and right eyes. Ideally, 2D stereo image pairs should simulate the same horizontal offset that exists between the eyes - approximately 6.5 cm.

     There are an almost unlimited number of ways to produce and view stereo images. Four techniques in particular are commonly used in conjunction with computers to visualize 3D models. The first uses red/blue (anaglyph) glasses to view objects rendered with red and blue highlights. Advantages of this technique include low cost (approximately 40 cents per pair of anaglyph glasses) and the simplicity of the technique. No specialized display systems are required. A disadvantage of this technique is that since red and blue are used to filter the image for each eye, objects in the scene should not be entirely red or entirely blue because they would be seen by only one eye.

     The second technique uses polarized light to separate the views intended for each eye. The advantage of this technique is that it allows an object to be any color. Disadvantages include the need for special hardware to polarize the light. This hardware will be different depending on the display technique (computer or slides). Polarizing computer screens can also cost in excess of $2000 and polarized images cannot be displayed in hard copy format.

     The third technique makes use of shuttered LCD glasses. In this technique, the computer alternately renders an object from two perspectives in sync with the LCD shutters in the glasses. At high shutter rates, the object will have the illusion of depth. Advantages of this system include the ability to use any color object in the rendered scene. Disadvantages include cost. LCD glasses typically cost around $800 each while the emitter used to control the glasses costs an additional $200. Also, it is often necessary to completely redesign a graphical user interface if the use of LCD glasses was not built into the system from the start. Like polarized stereo images, images used in shuttered LCD systems cannot be displayed in hard copy format and still retain the illusion of depth.

     The final technique is to use Virtual Reality to provide stereoscopic vision to the user. Virtual reality systems use two small video screens inside a head mounted display to provide views of a scene from two different perspectives. If coupled with a 3D tracking system, a user can walk around and interact with the 3D models. Major disadvantages of these systems include their very high cost and inability to provide stereoscopic display to users not wearing the head mounted display.

     The low cost of anaglyph glasses, the ease with which the rendering technique could be incorporated into an existing software system and the wide variety of display mediums available made the choice of red/blue stereoscopy the ideal one for our application.


Next: Methods Up: Title Page Previous: Full Text Index Index: Full Text Index Contents: Conference Page