A GPU based saliency map for high-fidelity selective rendering
AFRIGRAPH '06 Proceedings of the 4th international conference on Computer graphics, virtual reality, visualisation and interaction in Africa
A gaze-based study for investigating the perception of visual realism in simulated scenes
ACM Transactions on Applied Perception (TAP)
| Hi-index | 0.00 |
Even with modern graphics hardware, it is still notpossible to achieve high fidelity renderings of complexscenes in real time. However, as these images areproduced for human observers we may exploit the factthat the human eye is good, but not that good. Inparticular, it may be possible to render parts of animage at high quality and the rest of the scene at lowerquality without the user being aware of this difference.Image quality assessment algorithms, such as the Dalymodel, provide a measure of the perceptual qualitydifference between image pairs. This paper presents apsychophysical evaluation of an image quality metricand investigates how such models can be developed torapidly determine the parts of the scene with the mostnoticeable perceptual difference.