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

Computational models for the topological selectivity in early and primitive vision systems

Open Access Open Access

Abstract

A rich body of work in the cognitive sciences shows that humans are able to perceive topological properties at a basic level (short time presentation, apparent motion). Minsky and Papert proved that the perceptron cannot compute topological primitives like connectedness. This has been used as an argument against computational approaches to early vision (L. Chen, 1982, Science, 218, 699-700). A recent finding by Lettvin (personal communication, 1991) provides a further challenge to computer vision scientists. Lettvin found that the “bug detector” does not respond to 3 dark blobs, unless they are connected with 2 lines.

© 1993 Optical Society of America

PDF Article
More Like This
ISETBIO: Computational tools for modeling early human vision

David Brainard, Haomiao Jiang, Nicolas P. Cottaris, Fred Rieke, E.J. Chichilnisky, Joyce E. Farrell, and Brian A. Wandell
IT4A.4 Imaging Systems and Applications (IS) 2015

Experiments with a spatiochromatic model of early vision

Ron Gershon and John K. Tsotsos
WD2 OSA Annual Meeting (FIO) 1985

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.