Visual importance-biased image synthesis animation

Present ray tracing algorithms are computationally intensive,\udrequiring hours of computing time for complex scenes. Our\udprevious work has dealt with the development of an overall\udapproach to the application of visual attention to progressive\udand adaptive ray-tracing techniques. The approach facilitates\udlarge computational savings by modulating the supersampling\udrates in an image by the visual importance of the region being\udrendered.\udThis paper extends the approach by incorporating temporal\udchanges into the models and techniques developed, as it is\udexpected that further efficiency savings can be reaped for\udanimated scenes. Applications for this approach include\udentertainment, visualisation and simulation