Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Energy computations in motion and texture

Not Accessible

Your library or personal account may give you access

Abstract

Energy is a phase-independent measure of stimulus strength. We show that an energy computation (that is, the energy of the spatiotemporal spectrum of some possibly nonlinear stimulus transformation) determines perceptual strength in three domains: (1) motion that is carried by luminance (first-order motion), (2) motion that is carried by textural properties (second-order motion), and (3) texture orientation in ambiguous spatial displays. We explore the hypothesis that perceptual strength in all three domains is governed by the same energy algorithm; it is merely applied to different transformations or to different cross-sections of a spatiotemporal stimulus. We generate ambiguous stimuli with competing motion paths (or competing texture orientations) and measure the psychometric functions that describe the perceptual dominance of one of the stimulus components as a function of its contrast. These measurements are repeated at different contrast levels of the competing component. From these data, we infer (a) the nature of the strength computation and (b) the exponent of the power function that describes the growth of equivalent internal noise with stimulus contrast. We find that all three domains are served by energy computations. The exponents of the noise-growth power functions are strikingly similar within an observer in all three domains, although exponents for different observers differ widely. We conclude that similar energy algorithms govern first-and second-order motion strength in space-time and orientation strength in a purely spatial domain.

© 1992 Optical Society of America

PDF Article
More Like This
A Versatile Test of Motion Perception for the Macintosh II Computer

Joanne M. Wood, Mark A. Bullimore, and Kirk H. Swenson
TuE1 Noninvasive Assessment of the Visual System (NAVS) 1992

Visual texture segmentation based on oriented energy pyramids

James R. Bergen
FD1 OSA Annual Meeting (FIO) 1988

Against a “covariance metric” for motion perception: chromatic and luminance information do not combine

Andrei Gorea, Ilona Kovacs, and Thomas V. Papathomas
FOO1 OSA Annual Meeting (FIO) 1992

Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.