Real-time semantic context labeling for image understanding

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

1 Citation (Scopus)
1 Downloads (Pure)

Abstract

The use of context information in a scene is an important aid for full semantic scene understanding in security and surveillance applications. To this end, this paper presents an innovative semantic context-labeling algorithm for three context classes, trading-off quality and real-time execution. Our system consists of three consecutive stages: image segmentation, region-based feature extraction and classification. We propose the joint use of the features color in HSV space, texture from Gabor filters and spatial context, in combination with the Directional Nearest Neighbor (DNN) method for constructing the undirected graph for segmentation. Compared to recent literature, this combination is over 35 times faster and achieves a coverability rate that is 65% higher.
Original languageEnglish
Title of host publicationProceedings of the2015 IEEE International Conference on Image Processing (ICIP 2015), 27-30 September 2015, Quebec City, Canada
Place of PublicationPiscataway
PublisherInstitute of Electrical and Electronics Engineers
Pages3180-3184
ISBN (Print)978-1-4799-8339-1
DOIs
Publication statusPublished - 2015
Event22nd IEEE International Conference on Image Processing (ICIP 2015) - Quebec, Canada
Duration: 27 Sep 201530 Sep 2015
Conference number: 22
http://www.icip2015.org/

Conference

Conference22nd IEEE International Conference on Image Processing (ICIP 2015)
Abbreviated titleICIP 2015
CountryCanada
CityQuebec
Period27/09/1530/09/15
Internet address

Fingerprint

Image understanding
Labeling
Semantics
Gabor filters
Image segmentation
Feature extraction
Textures
Color

Cite this

Pieck, M. A. R., Sommen, van der, F., Zinger, S., & With, de, P. H. N. (2015). Real-time semantic context labeling for image understanding. In Proceedings of the2015 IEEE International Conference on Image Processing (ICIP 2015), 27-30 September 2015, Quebec City, Canada (pp. 3180-3184). Piscataway: Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICIP.2015.7351390
Pieck, M.A.R. ; Sommen, van der, F. ; Zinger, S. ; With, de, P.H.N. / Real-time semantic context labeling for image understanding. Proceedings of the2015 IEEE International Conference on Image Processing (ICIP 2015), 27-30 September 2015, Quebec City, Canada. Piscataway : Institute of Electrical and Electronics Engineers, 2015. pp. 3180-3184
@inproceedings{77af2b5bfe8e4a51a60cd8ef951538a9,
title = "Real-time semantic context labeling for image understanding",
abstract = "The use of context information in a scene is an important aid for full semantic scene understanding in security and surveillance applications. To this end, this paper presents an innovative semantic context-labeling algorithm for three context classes, trading-off quality and real-time execution. Our system consists of three consecutive stages: image segmentation, region-based feature extraction and classification. We propose the joint use of the features color in HSV space, texture from Gabor filters and spatial context, in combination with the Directional Nearest Neighbor (DNN) method for constructing the undirected graph for segmentation. Compared to recent literature, this combination is over 35 times faster and achieves a coverability rate that is 65{\%} higher.",
author = "M.A.R. Pieck and {Sommen, van der}, F. and S. Zinger and {With, de}, P.H.N.",
year = "2015",
doi = "10.1109/ICIP.2015.7351390",
language = "English",
isbn = "978-1-4799-8339-1",
pages = "3180--3184",
booktitle = "Proceedings of the2015 IEEE International Conference on Image Processing (ICIP 2015), 27-30 September 2015, Quebec City, Canada",
publisher = "Institute of Electrical and Electronics Engineers",
address = "United States",

}

Pieck, MAR, Sommen, van der, F, Zinger, S & With, de, PHN 2015, Real-time semantic context labeling for image understanding. in Proceedings of the2015 IEEE International Conference on Image Processing (ICIP 2015), 27-30 September 2015, Quebec City, Canada. Institute of Electrical and Electronics Engineers, Piscataway, pp. 3180-3184, 22nd IEEE International Conference on Image Processing (ICIP 2015), Quebec, Canada, 27/09/15. https://doi.org/10.1109/ICIP.2015.7351390

Real-time semantic context labeling for image understanding. / Pieck, M.A.R.; Sommen, van der, F.; Zinger, S.; With, de, P.H.N.

Proceedings of the2015 IEEE International Conference on Image Processing (ICIP 2015), 27-30 September 2015, Quebec City, Canada. Piscataway : Institute of Electrical and Electronics Engineers, 2015. p. 3180-3184.

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - Real-time semantic context labeling for image understanding

AU - Pieck, M.A.R.

AU - Sommen, van der, F.

AU - Zinger, S.

AU - With, de, P.H.N.

PY - 2015

Y1 - 2015

N2 - The use of context information in a scene is an important aid for full semantic scene understanding in security and surveillance applications. To this end, this paper presents an innovative semantic context-labeling algorithm for three context classes, trading-off quality and real-time execution. Our system consists of three consecutive stages: image segmentation, region-based feature extraction and classification. We propose the joint use of the features color in HSV space, texture from Gabor filters and spatial context, in combination with the Directional Nearest Neighbor (DNN) method for constructing the undirected graph for segmentation. Compared to recent literature, this combination is over 35 times faster and achieves a coverability rate that is 65% higher.

AB - The use of context information in a scene is an important aid for full semantic scene understanding in security and surveillance applications. To this end, this paper presents an innovative semantic context-labeling algorithm for three context classes, trading-off quality and real-time execution. Our system consists of three consecutive stages: image segmentation, region-based feature extraction and classification. We propose the joint use of the features color in HSV space, texture from Gabor filters and spatial context, in combination with the Directional Nearest Neighbor (DNN) method for constructing the undirected graph for segmentation. Compared to recent literature, this combination is over 35 times faster and achieves a coverability rate that is 65% higher.

U2 - 10.1109/ICIP.2015.7351390

DO - 10.1109/ICIP.2015.7351390

M3 - Conference contribution

SN - 978-1-4799-8339-1

SP - 3180

EP - 3184

BT - Proceedings of the2015 IEEE International Conference on Image Processing (ICIP 2015), 27-30 September 2015, Quebec City, Canada

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

ER -

Pieck MAR, Sommen, van der F, Zinger S, With, de PHN. Real-time semantic context labeling for image understanding. In Proceedings of the2015 IEEE International Conference on Image Processing (ICIP 2015), 27-30 September 2015, Quebec City, Canada. Piscataway: Institute of Electrical and Electronics Engineers. 2015. p. 3180-3184 https://doi.org/10.1109/ICIP.2015.7351390