Wide-baseline image change detection

Ziggy Jones, Mike Brookes, Pier Luigi Dragotti, David Benton

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present a fully automated method for the detection of changes within a scene between a reference and a sample image whose viewing angles differ by up to 30°. We also describe an extension to the SIFT technique that allows extracted feature points to be matched over wider viewing angles. Matched correspondences between reference and sample images are used to construct a Delaunay triangulation and changes are detected by comparing triangles after affine compensation using a dense SIFT metric. False positives are reduced by using a novel technique introduced as local plane matching (LPM) to match mean-shift segments in unmatched areas using the homographies of local planes to compensate for perspective distortions. The method is shown to achieve pixel-level equal error rates of 5% at a 10° azimuth view angle difference.

Original languageEnglish
Title of host publication2014 IEEE International Conference on Image Processing, ICIP 2014
PublisherIEEE
Pages1589-1593
Number of pages5
ISBN (Electronic)9781479957514
DOIs
Publication statusPublished - 28 Jan 2014

Fingerprint

Triangulation
Pixels
Compensation and Redress

Keywords

  • Affine Compensation
  • Change Detection
  • Feature Points
  • Image Matching
  • Local Features
  • Segmentation
  • SIFT
  • Wide Baseline

Cite this

Jones, Z., Brookes, M., Dragotti, P. L., & Benton, D. (2014). Wide-baseline image change detection. In 2014 IEEE International Conference on Image Processing, ICIP 2014 (pp. 1589-1593). [7025318] IEEE. https://doi.org/10.1109/ICIP.2014.7025318
Jones, Ziggy ; Brookes, Mike ; Dragotti, Pier Luigi ; Benton, David. / Wide-baseline image change detection. 2014 IEEE International Conference on Image Processing, ICIP 2014. IEEE, 2014. pp. 1589-1593
@inproceedings{212e114eaac44eebbc533a28c3b2792e,
title = "Wide-baseline image change detection",
abstract = "We present a fully automated method for the detection of changes within a scene between a reference and a sample image whose viewing angles differ by up to 30°. We also describe an extension to the SIFT technique that allows extracted feature points to be matched over wider viewing angles. Matched correspondences between reference and sample images are used to construct a Delaunay triangulation and changes are detected by comparing triangles after affine compensation using a dense SIFT metric. False positives are reduced by using a novel technique introduced as local plane matching (LPM) to match mean-shift segments in unmatched areas using the homographies of local planes to compensate for perspective distortions. The method is shown to achieve pixel-level equal error rates of 5{\%} at a 10° azimuth view angle difference.",
keywords = "Affine Compensation, Change Detection, Feature Points, Image Matching, Local Features, Segmentation, SIFT, Wide Baseline",
author = "Ziggy Jones and Mike Brookes and Dragotti, {Pier Luigi} and David Benton",
year = "2014",
month = "1",
day = "28",
doi = "10.1109/ICIP.2014.7025318",
language = "English",
pages = "1589--1593",
booktitle = "2014 IEEE International Conference on Image Processing, ICIP 2014",
publisher = "IEEE",
address = "United States",

}

Jones, Z, Brookes, M, Dragotti, PL & Benton, D 2014, Wide-baseline image change detection. in 2014 IEEE International Conference on Image Processing, ICIP 2014., 7025318, IEEE, pp. 1589-1593. https://doi.org/10.1109/ICIP.2014.7025318

Wide-baseline image change detection. / Jones, Ziggy; Brookes, Mike; Dragotti, Pier Luigi; Benton, David.

2014 IEEE International Conference on Image Processing, ICIP 2014. IEEE, 2014. p. 1589-1593 7025318.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Wide-baseline image change detection

AU - Jones, Ziggy

AU - Brookes, Mike

AU - Dragotti, Pier Luigi

AU - Benton, David

PY - 2014/1/28

Y1 - 2014/1/28

N2 - We present a fully automated method for the detection of changes within a scene between a reference and a sample image whose viewing angles differ by up to 30°. We also describe an extension to the SIFT technique that allows extracted feature points to be matched over wider viewing angles. Matched correspondences between reference and sample images are used to construct a Delaunay triangulation and changes are detected by comparing triangles after affine compensation using a dense SIFT metric. False positives are reduced by using a novel technique introduced as local plane matching (LPM) to match mean-shift segments in unmatched areas using the homographies of local planes to compensate for perspective distortions. The method is shown to achieve pixel-level equal error rates of 5% at a 10° azimuth view angle difference.

AB - We present a fully automated method for the detection of changes within a scene between a reference and a sample image whose viewing angles differ by up to 30°. We also describe an extension to the SIFT technique that allows extracted feature points to be matched over wider viewing angles. Matched correspondences between reference and sample images are used to construct a Delaunay triangulation and changes are detected by comparing triangles after affine compensation using a dense SIFT metric. False positives are reduced by using a novel technique introduced as local plane matching (LPM) to match mean-shift segments in unmatched areas using the homographies of local planes to compensate for perspective distortions. The method is shown to achieve pixel-level equal error rates of 5% at a 10° azimuth view angle difference.

KW - Affine Compensation

KW - Change Detection

KW - Feature Points

KW - Image Matching

KW - Local Features

KW - Segmentation

KW - SIFT

KW - Wide Baseline

UR - http://www.scopus.com/inward/record.url?scp=84949926928&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2014.7025318

DO - 10.1109/ICIP.2014.7025318

M3 - Conference contribution

AN - SCOPUS:84949926928

SP - 1589

EP - 1593

BT - 2014 IEEE International Conference on Image Processing, ICIP 2014

PB - IEEE

ER -

Jones Z, Brookes M, Dragotti PL, Benton D. Wide-baseline image change detection. In 2014 IEEE International Conference on Image Processing, ICIP 2014. IEEE. 2014. p. 1589-1593. 7025318 https://doi.org/10.1109/ICIP.2014.7025318