GFOVGeometric Field of View
References in periodicals archive ?
We manipulated the GFOV of a stereoscopic display between the nonperspective case (parallel projection: 0[degrees] GFOV) and a GFOV of 100[degrees] while maintaining a constant FOV.
Across GFOV levels the range-ring plane of the display was scaled to maintain a constant image size, thus avoiding the confound of image-size changes with differing GFOVs.
Given that the perspective distortions involved depend on the difference between FOV and GFOV angles, this makes comparisons difficult.
The display GFOV was manipulated over four levels: 0[degrees] (parallel projection), 13.
The four levels of GFOV were 0[degrees] (parallel projection), 13.
The experimental trials were blocked over GFOV conditions; participants performed each GFOV block on separate days.
Figure 3 shows two stereo pairs of the display (veridical and 100[degrees] GFOVs), which comprised a blue range-ring plane, diameter 72 mm (of invariant size across GFOV conditions), centered within a blue wire-frame box that delimited the 3D volume of the display.
The diameter of the target symbol varied across GFOV conditions and symbol elevation, with the greatest range of diameters in the 100[degrees] GFOV condition.
With the exception of those symbol positions within the range-ring plane and those on the centerline of the display (the line from the COP passing through the centers of the front and rear faces of the wire-frame box and that of the range-ring plane), there was an effect on the relative positions of symbol, range rings, and wire-frame box attributable to GFOV condition.
In general, GFOV angles larger than veridical compress points in the display volume toward the centerline of the display for elevations below the range-ring plane.
The perspective projections for each GFOV were produced by placing the COP at varying distances from the display screen.
This can be accomplished by equating GFOV and DFOV (assuming that a spatially accurate model of the VE exists).