Visual Equivalence: towards a new standard for image fidelity.

Ganesh Ramanarayanan, James Ferwerda, Bruce Walter and Kavita Bala

Efficient, high-quality rendering of complex scenes is one of the

grand challenges of computer graphics. Perceptually-based rendering

addresses this challenge by taking advantage of the limits of

human vision. However, existing methods, based on predicting visible

image differences, are too conservative because many image

differences do not matter to human observers. In this paper, we

introduce visual equivalence, a new standard for image fidelity in

graphics. Images are visually equivalent if they convey the same

impressions of scene appearance, even if they are visibly different.

To understand this phenomenon, we conduct a series of experiments

that explore how object geometry, material, and illumination interact

to provide information about appearance, and we characterize

how two kinds of transformations on illumination maps (blurring

and warping) affect these appearance attributes. We then derive

metrics for predicting when images rendered with the transformed

maps will be visually equivalent to images rendered with the reference

map, and we run a confirmatory study to validate the predictive

power of these metrics across different geometries, materials,

and illumination maps. Finally, we show how these new visual

equivalence predictors can be used to improve the efficiency of two

rendering algorithms: Lightcuts and precomputed radiance transfer.

This work represents some promising first steps towards developing

perceptual metrics based on higher order aspects of visual coding.