Perceptually guided corrective splatting

One of the basic difficulties with interactive walkthroughs is the high qua lity rendering of object Surfaces with non-diffuse light scattering charact eristics. Since full ray tracing at interactive rates is usually, impossibl e, we render a precomputed global illumination solution using graphics hard ware and use remaining computational power to correct the appearance of non -diffuse objects on-the-fly. The question arises, how to obtain the best im age quality as perceived by a human observer within a limited amount of tim e for each frame. Me address this problem by enforcing corrective computati on for those non-diffuse objects that are selected using a computational mo del of visual attention. We consider both the saliency- and task-driven sel ection of those objects and benefit from the fact that shading artifacts of "unattended" objects are likely to remain unnoticed. Me use a hierarchical image-space sampling scheme to control ray tracing and splat the generated point samples. The resulting image converges progressively to a ray traced solution if the viewing parameters remain unchanged. Moreover, we use a sa mple cache to enhance visual appearance if the time budget for correction h as been too low for some frame. We check the validity of the cached samples using a novel criterion suited for non-diffuse surfaces and reproject vali d samples into the current view.