PERCEPTUAL REPRESENTATION OF 3D SURFACES

Ohio State University, Columbus OH

Investigators

Linked publications & trials

Abstract

DESCRIPTION (Applicant's Abstract): The research described in this proposal is designed to measure the precision and accuracy of observers' judgments of various surface properties and relations that could potentially be used for the perceptual representation of 3D shape. Observers will be asked to estimate the distances between visible surface points, and to compare their local orientations and curvatures. The proposed research has four major goals. One is to measure how different attributes of surface structure are perceptually scaled, and the extent to which their representations are consistent with one another. A second is to determine how higher order properties of differential structure, such as orientation and curvature, are perceptually parameterized into component dimensions. Can these components be selected arbitrarily, or are there privileged coordinate systems? A third goal of the research is to investigate the qualitative aspects of perceived surface structure. What defines identifiable parts or features on smoothly curved surfaces, and to what extent are they viewpoint invariant? Finally, additional experiments will also be performed to investigate how perceptual knowledge of spatial relations is used to guide motor actions, such as reaching to a target.