A feasibility test for perceptually adaptive level of detail rendering on desktop systems

• Authors: Derrick Parkhurst;Ernst Niebur
• Venue: APGV '04 Proceedings of the 1st Symposium on Applied perception in graphics and visualization
• Year: 2004

Level of detail (LOD) rendering techniques reduce the geometric complexity of 3D models, sacrificing visual rendering quality in order to increase frame rendering rates. Perceptually adaptive LOD rendering techniques take into account the characteristics of the human visual system to minimize visible artifacts attributable to the reduced LOD. While these techniques have been previously examined in the context of high-performance rendering systems, it is not clear whether the benefits will necessarily overcome the behavioral costs associated with a reduced LOD on ordinary desktop systems. To answer this question, two perceptually adaptive rendering techniques, one velocity-dependent and one gaze-contingent, were implemented in the Unreal TM rendering engine on a standard desktop computer and monitor. These techniques were evaluated in separate experiments where participants were required to perform a virtual search for a target object among distractor objects in a perceptually rendered virtual home interior using a mouse to rotate the viewport. In the first experiment, objects moving across the observer's field of view were rendered in less detail than stationary objects, taking advantage of the fact that visual sensitivity to the details of moving objects is substantially reduced. Reaction times to detect the target remained constant with decreasing detail, whereas reaction times to localize a target decreased. In a second experiment, an eye tracker was used to render objects at the point of gaze in more detail than objects in the periphery, taking advantage of the fact that visual sensitivity is greatest at that location. Reaction times to detect the target increased with decreasing detail, whereas reaction times to localize a target decreased. The results from these experiments suggest that a reduced LOD can impede target identification, however, the resultant increase in frame rates facilitates virtual interaction. Overall, the behavioral costs associated with perceptually adaptive LOD techniques can be offset by the behavioral performance gains on desktop systems. However, we show that the nature of the task is important in determining the exact cost-benefit trade-off.