In this work, we provide some common methods, techniques, information, concepts, and relevant citations for those conducting human factors-related research with stereoscopic 3D (S3D) displays. We give suggested methods for calculating binocular disparities, and show how to verify on-screen image separation measurements. We provide typical values for inter-pupillary distances that are useful in such calculations. We discuss the pros, cons, and suggested uses of some common stereovision clinical tests. We discuss the phenomena and prevalence rates of stereoanomalous, pseudo-stereoanomalous, stereo-deficient, and stereoblind viewers. The problems of eyestrain and fatigue-related effects from stereo viewing, and the possible causes, are enumerated. System and viewer crosstalk are defined and discussed, and the issue of stereo camera separation is explored. Typical binocular fusion limits are also provided for reference, and discussed in relation to zones of comfort. Finally, the concept of measuring disparity distributions is described. The implications of these issues for the human factors study of S3D displays are covered throughout.
INTRODUCTIONOver the last decade, stereoscopic 3D (S3D) displays have reached the mass markets and are now finding interest and application in a variety of fields, including in entertainment, industrial design, medicine, and the military. The growing popularity of S3D technology stems primarily from its ability to provide a viewer with binocular cues to depth in a scene (stereopsis), a capability that no other mass-market display can currently provide. Researchers, engineers, display designers, and visualization developers now have the sensors, computer graphics tools, interaction devices, and displays to (potentially) take full advantage of the third dimension.But despite this embarrassment of riches in terms of data, tools, and displays, characterizing and studying S3D systems is by no means a straightforward endeavor. This is complicated by the fact that many of the traditional experimental tools, techniques, methods, rules-of-thumb, and ways of thinking about display evaluation are sometimes inadequate (or even inappropriate) for these new 3D systems. In this work, we provide a handy, reference 'field guide' for those interested in performing human-factors psychology/engineering studies and evaluations of S3D systems, in the hopes of easing the way forward for widespread 3D display research, development, and use (see Hsu, Pizlo, Chelberg, Babbs, & Delp, 1996, for a brief guide for experimentally comparing non-stereo 2D versus stereoscopic 3D).