Training-free moving object detection system based on hierarchical color-guided motion segmentation

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

1 Citation (Scopus)

Abstract

We present a moving object detection system for surveillance based on Hierarchical Color-guided Motion segmentation (HiCoMo). The HiCoMo system does not require training and consists of two main stages: (1) hierarchical color-guided motion segmentation, and (2) motion-based verification. The first stage is a hierarchical segmentation framework, where at each level a balance is made between static and temporal features. So that groups of pixels develop into semantic object segments. In the second stage, these object segments are further analyzed in terms of motion saliency and consistency, in order to finalize the object detection results. Our proposed system is tested on real-life surveillance videos containing various scenarios. The detection results outperform a state-of-the-art training-free moving object detection algorithm in recall (90.2% compared to 81.6%) while having a competitively promising precision (96.5% compared to 97.4%). The system has a generic nature and real-time implementation potential, which makes it applicable to various applications of computer vision.

LanguageEnglish
Title of host publicationProceedings of the 14th IAPR International Conference on Machine Vision Applications, MVA 2015
Place of PublicationPiscataway
PublisherInstitute of Electrical and Electronics Engineers
Pages154-157
Number of pages4
ISBN (Electronic)9784901122153
DOIs
StatePublished - 1 Jan 2015
Event14th IAPR International Conference on Machine Vision Applications (MVA 2015) - Tokyo, Japan
Duration: 18 May 201522 May 2015
Conference number: 14
http://www.mva-org.jp/mva2015

Conference

Conference14th IAPR International Conference on Machine Vision Applications (MVA 2015)
Abbreviated titleMVA 2015
CountryJapan
CityTokyo
Period18/05/1522/05/15
Internet address

Fingerprint

Color
Computer vision
Pixels
Semantics
Object detection

Cite this

Bao, X., Dubbelman, G., Zinger, S., & de With, P. H. N. (2015). Training-free moving object detection system based on hierarchical color-guided motion segmentation. In Proceedings of the 14th IAPR International Conference on Machine Vision Applications, MVA 2015 (pp. 154-157). Piscataway: Institute of Electrical and Electronics Engineers. DOI: 10.1109/MVA.2015.7153156
Bao, Xinfeng ; Dubbelman, Gijs ; Zinger, Svitlana ; de With, Peter H.N./ Training-free moving object detection system based on hierarchical color-guided motion segmentation. Proceedings of the 14th IAPR International Conference on Machine Vision Applications, MVA 2015. Piscataway : Institute of Electrical and Electronics Engineers, 2015. pp. 154-157
@inproceedings{f8cbf7488fc7474794f598d8f80b61bd,
title = "Training-free moving object detection system based on hierarchical color-guided motion segmentation",
abstract = "We present a moving object detection system for surveillance based on Hierarchical Color-guided Motion segmentation (HiCoMo). The HiCoMo system does not require training and consists of two main stages: (1) hierarchical color-guided motion segmentation, and (2) motion-based verification. The first stage is a hierarchical segmentation framework, where at each level a balance is made between static and temporal features. So that groups of pixels develop into semantic object segments. In the second stage, these object segments are further analyzed in terms of motion saliency and consistency, in order to finalize the object detection results. Our proposed system is tested on real-life surveillance videos containing various scenarios. The detection results outperform a state-of-the-art training-free moving object detection algorithm in recall (90.2{\%} compared to 81.6{\%}) while having a competitively promising precision (96.5{\%} compared to 97.4{\%}). The system has a generic nature and real-time implementation potential, which makes it applicable to various applications of computer vision.",
author = "Xinfeng Bao and Gijs Dubbelman and Svitlana Zinger and {de With}, {Peter H.N.}",
year = "2015",
month = "1",
day = "1",
doi = "10.1109/MVA.2015.7153156",
language = "English",
pages = "154--157",
booktitle = "Proceedings of the 14th IAPR International Conference on Machine Vision Applications, MVA 2015",
publisher = "Institute of Electrical and Electronics Engineers",
address = "United States",

}

Bao, X, Dubbelman, G, Zinger, S & de With, PHN 2015, Training-free moving object detection system based on hierarchical color-guided motion segmentation. in Proceedings of the 14th IAPR International Conference on Machine Vision Applications, MVA 2015. Institute of Electrical and Electronics Engineers, Piscataway, pp. 154-157, 14th IAPR International Conference on Machine Vision Applications (MVA 2015), Tokyo, Japan, 18/05/15. DOI: 10.1109/MVA.2015.7153156

Training-free moving object detection system based on hierarchical color-guided motion segmentation. / Bao, Xinfeng; Dubbelman, Gijs; Zinger, Svitlana; de With, Peter H.N.

Proceedings of the 14th IAPR International Conference on Machine Vision Applications, MVA 2015. Piscataway : Institute of Electrical and Electronics Engineers, 2015. p. 154-157.

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

TY - GEN

T1 - Training-free moving object detection system based on hierarchical color-guided motion segmentation

AU - Bao,Xinfeng

AU - Dubbelman,Gijs

AU - Zinger,Svitlana

AU - de With,Peter H.N.

PY - 2015/1/1

Y1 - 2015/1/1

N2 - We present a moving object detection system for surveillance based on Hierarchical Color-guided Motion segmentation (HiCoMo). The HiCoMo system does not require training and consists of two main stages: (1) hierarchical color-guided motion segmentation, and (2) motion-based verification. The first stage is a hierarchical segmentation framework, where at each level a balance is made between static and temporal features. So that groups of pixels develop into semantic object segments. In the second stage, these object segments are further analyzed in terms of motion saliency and consistency, in order to finalize the object detection results. Our proposed system is tested on real-life surveillance videos containing various scenarios. The detection results outperform a state-of-the-art training-free moving object detection algorithm in recall (90.2% compared to 81.6%) while having a competitively promising precision (96.5% compared to 97.4%). The system has a generic nature and real-time implementation potential, which makes it applicable to various applications of computer vision.

AB - We present a moving object detection system for surveillance based on Hierarchical Color-guided Motion segmentation (HiCoMo). The HiCoMo system does not require training and consists of two main stages: (1) hierarchical color-guided motion segmentation, and (2) motion-based verification. The first stage is a hierarchical segmentation framework, where at each level a balance is made between static and temporal features. So that groups of pixels develop into semantic object segments. In the second stage, these object segments are further analyzed in terms of motion saliency and consistency, in order to finalize the object detection results. Our proposed system is tested on real-life surveillance videos containing various scenarios. The detection results outperform a state-of-the-art training-free moving object detection algorithm in recall (90.2% compared to 81.6%) while having a competitively promising precision (96.5% compared to 97.4%). The system has a generic nature and real-time implementation potential, which makes it applicable to various applications of computer vision.

UR - http://www.scopus.com/inward/record.url?scp=84941243835&partnerID=8YFLogxK

U2 - 10.1109/MVA.2015.7153156

DO - 10.1109/MVA.2015.7153156

M3 - Conference contribution

SP - 154

EP - 157

BT - Proceedings of the 14th IAPR International Conference on Machine Vision Applications, MVA 2015

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

ER -

Bao X, Dubbelman G, Zinger S, de With PHN. Training-free moving object detection system based on hierarchical color-guided motion segmentation. In Proceedings of the 14th IAPR International Conference on Machine Vision Applications, MVA 2015. Piscataway: Institute of Electrical and Electronics Engineers. 2015. p. 154-157. Available from, DOI: 10.1109/MVA.2015.7153156