Key-point Detection with Multi-layer Center-surround Inhibition

Foti Coleca; Sabrina Zîrnovean; Thomas Käster; Thomas Martinetz; Erhardt Barth

doi:10.5220/0004743103860393

Key-point Detection with Multi-layer Center-surround Inhibition

Foti Coleca, Sabrina Zîrnovean, Thomas Käster, Thomas Martinetz, Erhardt Barth

Institute of Neuro- and Bioinformatics

University POLITEHNICA of Bucureşti

Abstract

We present a biologically inspired algorithm for key-point detection based on multi-layer and nonlinear center-surround inhibition. A Bag-of-Visual-Words framework is used to evaluate the performance of the detector on the Oxford III-T Pet Dataset for pet recognition. The results demonstrate an increased performance of our algorithm compared to the SIFT key-point detector. We further improve the recognition rate by separately training codebooks for the ON- and OFF-type key points. The results show that our key-point detection algorithms outperform the SIFT detector by having a lower recognition-error rate over a whole range of different key-point densities. Randomly selected keypoints are also outperformed.

Original language	English
Title of host publication	Proceedings of the 9th International Conference on Computer Vision Theory and Application
Editors	Sebastiano Battiato
Number of pages	8
Volume	Vol. 1
Place of Publication	Lisbon, Portugal
Publisher	SciTePress
Publication date	01.01.2014
Edition	1
Pages	386-393
ISBN (Print)	9897580042, 978-9897580048
DOIs	https://doi.org/10.5220/0004743103860393
Publication status	Published - 01.01.2014
Event	The International Conference on Computer Vision Theory and Applications - Lisbon / Lissabon, Portugal Duration: 05.01.2014 → 08.01.2014 http://www.visapp.visigrapp.org/?y=2014

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being
SDG 9 Industry, Innovation, and Infrastructure

Access to Document

10.5220/0004743103860393

Cite this

Coleca, F., Zîrnovean, S., Käster, T., Martinetz, T., & Barth, E. (2014). Key-point Detection with Multi-layer Center-surround Inhibition. In S. Battiato (Ed.), Proceedings of the 9th International Conference on Computer Vision Theory and Application (1 ed., Vol. Vol. 1, pp. 386-393). (INSTICC Conference Proceedings). SciTePress. https://doi.org/10.5220/0004743103860393

@inproceedings{3a1738ad0d8c4c4087a04411fda50e78,

title = "Key-point Detection with Multi-layer Center-surround Inhibition",

abstract = "We present a biologically inspired algorithm for key-point detection based on multi-layer and nonlinear center-surround inhibition. A Bag-of-Visual-Words framework is used to evaluate the performance of the detector on the Oxford III-T Pet Dataset for pet recognition. The results demonstrate an increased performance of our algorithm compared to the SIFT key-point detector. We further improve the recognition rate by separately training codebooks for the ON- and OFF-type key points. The results show that our key-point detection algorithms outperform the SIFT detector by having a lower recognition-error rate over a whole range of different key-point densities. Randomly selected keypoints are also outperformed.",

author = "Foti Coleca and Sabrina Z{\^i}rnovean and Thomas K{\"a}ster and Thomas Martinetz and Erhardt Barth",

year = "2014",

month = jan,

day = "1",

doi = "10.5220/0004743103860393",

language = "English",

isbn = "9897580042",

volume = "Vol. 1",

series = "INSTICC Conference Proceedings",

publisher = "SciTePress",

pages = "386--393",

editor = "Sebastiano Battiato",

booktitle = "Proceedings of the 9th International Conference on Computer Vision Theory and Application",

edition = "1",

note = "The International Conference on Computer Vision Theory and Applications ; Conference date: 05-01-2014 Through 08-01-2014",

url = "http://www.visapp.visigrapp.org/?y=2014",

}

Coleca, F, Zîrnovean, S, Käster, T , Martinetz, T & Barth, E 2014, Key-point Detection with Multi-layer Center-surround Inhibition. in S Battiato (ed.), Proceedings of the 9th International Conference on Computer Vision Theory and Application. 1 edn, vol. Vol. 1, INSTICC Conference Proceedings, SciTePress, Lisbon, Portugal, pp. 386-393, The International Conference on Computer Vision Theory and Applications , Lisbon / Lissabon, Portugal, 05.01.14. https://doi.org/10.5220/0004743103860393

Key-point Detection with Multi-layer Center-surround Inhibition. / Coleca, Foti; Zîrnovean, Sabrina; Käster, Thomas et al.
Proceedings of the 9th International Conference on Computer Vision Theory and Application. ed. / Sebastiano Battiato. Vol. Vol. 1 1. ed. Lisbon, Portugal: SciTePress, 2014. p. 386-393 (INSTICC Conference Proceedings).

Research output: Chapters in Books/Reports/Conference Proceedings › Conference contribution › peer-review

TY - GEN

T1 - Key-point Detection with Multi-layer Center-surround Inhibition

AU - Coleca, Foti

AU - Zîrnovean, Sabrina

AU - Käster, Thomas

AU - Martinetz, Thomas

AU - Barth, Erhardt

PY - 2014/1/1

Y1 - 2014/1/1

N2 - We present a biologically inspired algorithm for key-point detection based on multi-layer and nonlinear center-surround inhibition. A Bag-of-Visual-Words framework is used to evaluate the performance of the detector on the Oxford III-T Pet Dataset for pet recognition. The results demonstrate an increased performance of our algorithm compared to the SIFT key-point detector. We further improve the recognition rate by separately training codebooks for the ON- and OFF-type key points. The results show that our key-point detection algorithms outperform the SIFT detector by having a lower recognition-error rate over a whole range of different key-point densities. Randomly selected keypoints are also outperformed.

AB - We present a biologically inspired algorithm for key-point detection based on multi-layer and nonlinear center-surround inhibition. A Bag-of-Visual-Words framework is used to evaluate the performance of the detector on the Oxford III-T Pet Dataset for pet recognition. The results demonstrate an increased performance of our algorithm compared to the SIFT key-point detector. We further improve the recognition rate by separately training codebooks for the ON- and OFF-type key points. The results show that our key-point detection algorithms outperform the SIFT detector by having a lower recognition-error rate over a whole range of different key-point densities. Randomly selected keypoints are also outperformed.

UR - http://www.scitepress.org/PublicationsDetail.aspx?ID=njTfDvfKjDs=&t=1

U2 - 10.5220/0004743103860393

DO - 10.5220/0004743103860393

M3 - Conference contribution

SN - 9897580042

SN - 978-9897580048

VL - Vol. 1

T3 - INSTICC Conference Proceedings

SP - 386

EP - 393

BT - Proceedings of the 9th International Conference on Computer Vision Theory and Application

A2 - Battiato, Sebastiano

PB - SciTePress

CY - Lisbon, Portugal

T2 - The International Conference on Computer Vision Theory and Applications

Y2 - 5 January 2014 through 8 January 2014

ER -

Key-point Detection with Multi-layer Center-surround Inhibition

Abstract

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this