The University of Southampton
University of Southampton Institutional Repository

Symmetry similarity of human perception to computer vision operators

Symmetry similarity of human perception to computer vision operators
Symmetry similarity of human perception to computer vision operators
Symmetry occurs everywhere around us and is key to human visual perception. Human perception can help guide the improvement of computer vision operators and this is the first paper aiming to quantify that guidance. We define the Degree of Symmetry (DoS) as the measure of ‘how symmetrical’ a region is as human perception sees symmetry in a continuous manner. A new dataset of symmetry axes, the Degree of Symmetry Axis Set, is compiled for ordering by DoS. A human perception rank order is found by crowd-sourced pairwise comparisons. The correlation of two ranked orders is defined as the Symmetry Similarity which we use to evaluate symmetry operators against human perception. No existing symmetry operator gives a value for DoS of a reflection axis. We extend three operators to give a value for the DoS of an axis: the Generalised Symmetry Transform, Loy’s interest point operator, and Griffin’s Derivative-of-Gaussian operator. The highest Symmetry Similarity of a symmetry operator to human perception revealed they are a poor approximation of human perception of symmetry.
Forrest, P.
44b92a54-79d0-4070-abbc-a8da45cb9994
Nixon, M.S.
2b5b9804-5a81-462a-82e6-92ee5fa74e12
Forrest, P.
44b92a54-79d0-4070-abbc-a8da45cb9994
Nixon, M.S.
2b5b9804-5a81-462a-82e6-92ee5fa74e12

Forrest, P. and Nixon, M.S. (2015) Symmetry similarity of human perception to computer vision operators. Advances in Visual Computing.

Record type: Conference or Workshop Item (Paper)

Abstract

Symmetry occurs everywhere around us and is key to human visual perception. Human perception can help guide the improvement of computer vision operators and this is the first paper aiming to quantify that guidance. We define the Degree of Symmetry (DoS) as the measure of ‘how symmetrical’ a region is as human perception sees symmetry in a continuous manner. A new dataset of symmetry axes, the Degree of Symmetry Axis Set, is compiled for ordering by DoS. A human perception rank order is found by crowd-sourced pairwise comparisons. The correlation of two ranked orders is defined as the Symmetry Similarity which we use to evaluate symmetry operators against human perception. No existing symmetry operator gives a value for DoS of a reflection axis. We extend three operators to give a value for the DoS of an axis: the Generalised Symmetry Transform, Loy’s interest point operator, and Griffin’s Derivative-of-Gaussian operator. The highest Symmetry Similarity of a symmetry operator to human perception revealed they are a poor approximation of human perception of symmetry.

Full text not available from this repository.

More information

Published date: December 2015
Venue - Dates: Advances in Visual Computing, 2015-12-01
Organisations: Vision, Learning and Control

Identifiers

Local EPrints ID: 385632
URI: http://eprints.soton.ac.uk/id/eprint/385632
PURE UUID: 3d045079-761b-4ebf-9487-c7d7ae3ef79a
ORCID for M.S. Nixon: ORCID iD orcid.org/0000-0002-9174-5934

Catalogue record

Date deposited: 07 Jan 2016 17:18
Last modified: 07 Oct 2020 02:38

Export record

Download statistics

Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.

View more statistics

Atom RSS 1.0 RSS 2.0

Contact ePrints Soton: eprints@soton.ac.uk

ePrints Soton supports OAI 2.0 with a base URL of http://eprints.soton.ac.uk/cgi/oai2

This repository has been built using EPrints software, developed at the University of Southampton, but available to everyone to use.

We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we will assume that you are happy to receive cookies on the University of Southampton website.

×