Towards Attention-aware Foveated RenderingTowards Attention-aware Foveated Rendering - 2302.01368.pdf

Tags

Top Highlights

• eccentricity-dependent aspects of human vision

• attention-aware model of contrast sensitivity

• Visual attention refers to a set of cognitive operations that helps us selectively process the vast amounts of information with which we are confronted

• Most often, we direct our attention overtly, by moving our eyes towards a location, but we can also direct attention to an area in the periphery covertly

• account for the effect of covert at- tention when modeling human contrast sensitivity

• modulating the amount of attention allo- cated to the contrast discrimination task in the periphery, by forcing attention to the fovea with a visually demanding task

• eccentricity-dependent effects of attention on contrast dis-

• We introduce the first analytic model of contrast sensitivity

• bandwidth considerations and demonstrate that our model may afford significant bandwidth savings

• do not attempt to derive a measurement instrument for attention allocation across the visual field

• nor do we propose new foveated rendering algorithms or specific compression

• such as acuity

• minimize the bandwidth of a graphics system

• none of them model cognitive or attentional effects of human vision

• which we aim to address with our work

• acuity can be understood as the outer limit of the CSF gamut where the sensitivity of vision drops sharply

• this effect has not been modeled over the visual field, nor is there any available data for the effect in the periphery (> 10◦), which we hope to rectify with our work

• modulating the amount of attention has been shown to affect contrast discrimination thresholds (see Sec. 2.3),

• spatio-temporal contrast sensitivity function (CSF),

• the inverse of the contrast discrimination threshold, that is, the smallest contrast of sinusoidal grating that can be perceived at each spatial and temporal frequency