Learning salient structures for the analysis of symmetric patterns
Learning salient structures for the analysis of symmetric patterns
Feature-based symmetry detection algorithms have become popular amongst researchers due to their dominance in performance, nevertheless, these approaches are computationally demanding. Also they are reliant on the presence of matched features, therefore they benefit from the abundance of detected keypoints; this implies that a trade-off between performance and computation time must be found. In this paper both issues are addressed, the detection of large sets of keypoints and the computation time for feature-based symmetry detection algorithms. We present an innovative process to learn rotation-invariant salient structures by clustering self-similarities. Keypoints are detected as local maxima in feature-maps computed using the learnt structures. Keypoints are described using BRISK. We consider an axis of symmetry to be a dense cloud of points in a parameter-space, a density-based clustering algorithm is used to find such clouds. Computing times are drastically shortened taking an average of 0.619 seconds to process an image. Detection results for single and multiple, straight and curved, reflection and glide-reflection symmetries are similar to the current state of the art.
286-295
Lomeli-R., Jaime
26137049-8361-4a68-a7d9-734befbc5e58
Nixon, Mark
2b5b9804-5a81-462a-82e6-92ee5fa74e12
5 July 2017
Lomeli-R., Jaime
26137049-8361-4a68-a7d9-734befbc5e58
Nixon, Mark
2b5b9804-5a81-462a-82e6-92ee5fa74e12
Lomeli-R., Jaime and Nixon, Mark
(2017)
Learning salient structures for the analysis of symmetric patterns.
Karray, F., Campilho, A. and Cheriet, F.
(eds.)
In Image Analysis and Recognition. ICIAR 2017.
vol. 10317,
Springer.
.
(doi:10.1007/978-3-319-59876-5_32).
Record type:
Conference or Workshop Item
(Paper)
Abstract
Feature-based symmetry detection algorithms have become popular amongst researchers due to their dominance in performance, nevertheless, these approaches are computationally demanding. Also they are reliant on the presence of matched features, therefore they benefit from the abundance of detected keypoints; this implies that a trade-off between performance and computation time must be found. In this paper both issues are addressed, the detection of large sets of keypoints and the computation time for feature-based symmetry detection algorithms. We present an innovative process to learn rotation-invariant salient structures by clustering self-similarities. Keypoints are detected as local maxima in feature-maps computed using the learnt structures. Keypoints are described using BRISK. We consider an axis of symmetry to be a dense cloud of points in a parameter-space, a density-based clustering algorithm is used to find such clouds. Computing times are drastically shortened taking an average of 0.619 seconds to process an image. Detection results for single and multiple, straight and curved, reflection and glide-reflection symmetries are similar to the current state of the art.
Text
Learning Salient Structures for the Analysis of Symmetric Patterns
- Accepted Manuscript
More information
Accepted/In Press date: 13 March 2017
e-pub ahead of print date: 2 June 2017
Published date: 5 July 2017
Venue - Dates:
14th International Conference on Image Analysis and Recognition, Polytechnique Montréal, Montreal, Canada, 2017-07-05 - 2017-07-07
Organisations:
Vision, Learning and Control, Electronics & Computer Science
Identifiers
Local EPrints ID: 408138
URI: http://eprints.soton.ac.uk/id/eprint/408138
ISSN: 0302-9743
PURE UUID: 1dd9b150-694d-4088-a6ee-3e888989d70a
Catalogue record
Date deposited: 12 May 2017 04:04
Last modified: 16 Mar 2024 05:38
Export record
Altmetrics
Contributors
Author:
Jaime Lomeli-R.
Editor:
F. Karray
Editor:
A. Campilho
Editor:
F. Cheriet
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
