vix.ing · top · new · best · stats · spec

A quantitative analysis of tilt in the Caf 'e Wall illusion: a\n bioplausible model for foveal and peripheral vision

2016/09/22 by Nasim Nematzadeh, Nematzadeh, Nasim, David Powers +1 · 1 citation
Neuroscience · Medicine · #Visual perception and processing mechanisms #Retinal Imaging and Analysis

paper · pdf · doi:10.48550/arxiv.1609.06927

Abstract

The biological characteristics of human visual processing can be investigated\nthrough the study of optical illusions and their perception, giving rise to\nintuitions that may improve computer vision to match human performance.\nGeometric illusions are a specific subfamily in which orientations and angles\nare misperceived. This paper reports quantifiable predictions of the degree of\ntilt for a typical geometric illusion called Caf 'e Wall, in which the mortar\nbetween the tiles seems to tilt or bow. Our study employs a common bioplausible\nmodel of retinal processing and we further develop an analytic processing\npipeline to quantify and thus predict the specific angle of tilt. We further\nstudy the effect of resolution and feature size in order to predict the\ndifferent perceived tilts in different areas of the fovea and periphery, where\nresolution varies as the eye saccades to different parts of the image. In the\nexperiments, several different minimal portions of the pattern, modeling\nmonocular and binocular foveal views, are investigated across multiple scales,\nin order to quantify tilts with confidence intervals and explore the difference\nbetween local and global tilt.\n

Cited by

Related