Quality Evaluation of Projection-Based VR Displays

From EuroVR Knowledge Base

Description

The goal of the tests described here is to be able to quickly determine whether there is an error, and what the source of that error is likely to be, without a lot of special equipment or complicated procedures. The tests described here are solely concerned with evaluating the visual portion of a VR system, that is, viewer-centered, stereoscopic images. Tracking position errors can be checked by looking at straight edges. Position an object so that one or more edges cross screen boundaries, and look at them through the tracked glasses. If the tracking is accurate, the edges will appear straight. If it is inaccurate, they will appear bent; in that case, the scale of the error can be estimated by removing the tracked glasses and moving them until the edges do look straight (from the now un-tracked viewpoint). The stereo phase can be checked by looking through the glasses one eye at a time. Overall latency in the tracking and rendering can be judged by observing a virtual object that has its position attached to the wand. Simply moving the wand and seeing how much the object lags behind gives a general sense

Primary Reference or Source

• Dave Pape, Dan Sandin, Quality Evaluation of Projection-Based VR Displays, Electronic Visualization Laboratory University of Illinois at Chicago, IPT 2000, June 2000.

Scope

Item: Display Systems

Properties:

• Accuracy
  • Accuracy of tracking
  • Stereo phase

• Efficiency
  • Overall latency in the tracking and rendering