Are performance differences of interest operators statistically significant?
Are performance differences of interest operators statistically significant?
The differences in performance of a range of interest operators are examined in a null hypothesis framework using McNemar’s test on a widely-used database of images, to ascertain whether these apparent differences are statistically significant. It is found that some performance differences are indeed statistically significant, though most of them are at a fairly low level of confidence, i.e. with about a 1-in-20 chance that the results could be due to features of the evaluation database. A new evaluation measure i.e. accurate homography estimation is used to characterize the performance of feature extraction algorithms.Results suggest that operators employing longer descriptors are more reliable.
Feature Extraction, Homography, McNemar's Test
429-436
Kanwal, Nadia
636e56ff-8bec-4857-815d-1d458ee81d69
Ehsan, Shoaib
ae8922f0-dbe0-4b22-8474-98e84d852de7
Clark, Adrian F.
81c08359-a1fe-4380-adc0-2da681e19df0
2011
Kanwal, Nadia
636e56ff-8bec-4857-815d-1d458ee81d69
Ehsan, Shoaib
ae8922f0-dbe0-4b22-8474-98e84d852de7
Clark, Adrian F.
81c08359-a1fe-4380-adc0-2da681e19df0
Kanwal, Nadia, Ehsan, Shoaib and Clark, Adrian F.
(2011)
Are performance differences of interest operators statistically significant?
In,
Real, P, Diaz-Pernil, D, Molina-Abril, H, Berciano, A and Kropatsch, W
(eds.)
Computer analysis of images and patterns: CAIP 2011.
(Lecture Notes in Computer Science, 6855)
Springer, .
(doi:10.1007/978-3-642-23678-5_51).
Record type:
Book Section
Abstract
The differences in performance of a range of interest operators are examined in a null hypothesis framework using McNemar’s test on a widely-used database of images, to ascertain whether these apparent differences are statistically significant. It is found that some performance differences are indeed statistically significant, though most of them are at a fairly low level of confidence, i.e. with about a 1-in-20 chance that the results could be due to features of the evaluation database. A new evaluation measure i.e. accurate homography estimation is used to characterize the performance of feature extraction algorithms.Results suggest that operators employing longer descriptors are more reliable.
This record has no associated files available for download.
More information
Published date: 2011
Keywords:
Feature Extraction, Homography, McNemar's Test
Identifiers
Local EPrints ID: 478884
URI: http://eprints.soton.ac.uk/id/eprint/478884
PURE UUID: 5aa747a9-cf97-4123-9c78-d39481040e52
Catalogue record
Date deposited: 12 Jul 2023 16:38
Last modified: 17 Mar 2024 04:16
Export record
Altmetrics
Contributors
Author:
Nadia Kanwal
Author:
Shoaib Ehsan
Author:
Adrian F. Clark
Editor:
P Real
Editor:
D Diaz-Pernil
Editor:
H Molina-Abril
Editor:
A Berciano
Editor:
W Kropatsch
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