Multi-view Pedestrian Detection Using Statistical Colour Matching

Jie Ren; Ming Xu; Jeremy S. Smith

doi:10.1109/BigMM.2015.84

Multi-view Pedestrian Detection Using Statistical Colour Matching

Jie Ren, Ming Xu, Jeremy S. Smith

Department of Electrical and Electronic Engineering

Research output: Chapter in Book or Report/Conference proceeding › Conference Proceeding › peer-review

1 Citation (Scopus)

Abstract

To increase the robustness of detection in intelligent video surveillance systems, homography has been widely used to fuse foreground regions projected from multiple camera views to a reference view. The objective of this paper is to detect multiple pedestrians and identify the false-positive detections, which occur due to the foreground intersections of non-corresponding objects, in the top view using occupancy information and colour matching. Multiple homographies are used to detect the head plane and height of each pedestrian. The head locations can be used in the further tracking part. Experimental results show good performance of this method.

Original language	English
Title of host publication	Proceedings - 2015 IEEE International Conference on Multimedia Big Data, BigMM 2015
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	300-305
Number of pages	6
ISBN (Electronic)	9781479986880
DOIs	https://doi.org/10.1109/BigMM.2015.84
Publication status	Published - 9 Jul 2015
Event	1st IEEE International Conference on Multimedia Big Data, BigMM 2015 - Beijing, China Duration: 20 Apr 2015 → 22 Apr 2015

Publication series

Name	Proceedings - 2015 IEEE International Conference on Multimedia Big Data, BigMM 2015

Conference

Conference	1st IEEE International Conference on Multimedia Big Data, BigMM 2015
Country/Territory	China
City	Beijing
Period	20/04/15 → 22/04/15

Keywords

compressive sensing
sparse representation
structure set prediction
subspace learning
visual categorization

Access to Document

10.1109/BigMM.2015.84

Cite this

Ren, J., Xu, M., & Smith, J. S. (2015). Multi-view Pedestrian Detection Using Statistical Colour Matching. In Proceedings - 2015 IEEE International Conference on Multimedia Big Data, BigMM 2015 (pp. 300-305). Article 7153904 (Proceedings - 2015 IEEE International Conference on Multimedia Big Data, BigMM 2015). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BigMM.2015.84

@inproceedings{9045d48ceaf7427092c8713881a94b15,

title = "Multi-view Pedestrian Detection Using Statistical Colour Matching",

abstract = "To increase the robustness of detection in intelligent video surveillance systems, homography has been widely used to fuse foreground regions projected from multiple camera views to a reference view. The objective of this paper is to detect multiple pedestrians and identify the false-positive detections, which occur due to the foreground intersections of non-corresponding objects, in the top view using occupancy information and colour matching. Multiple homographies are used to detect the head plane and height of each pedestrian. The head locations can be used in the further tracking part. Experimental results show good performance of this method.",

keywords = "compressive sensing, sparse representation, structure set prediction, subspace learning, visual categorization",

author = "Jie Ren and Ming Xu and Smith, {Jeremy S.}",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; 1st IEEE International Conference on Multimedia Big Data, BigMM 2015 ; Conference date: 20-04-2015 Through 22-04-2015",

year = "2015",

month = jul,

day = "9",

doi = "10.1109/BigMM.2015.84",

language = "English",

series = "Proceedings - 2015 IEEE International Conference on Multimedia Big Data, BigMM 2015",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "300--305",

booktitle = "Proceedings - 2015 IEEE International Conference on Multimedia Big Data, BigMM 2015",

}

Ren, J, Xu, M & Smith, JS 2015, Multi-view Pedestrian Detection Using Statistical Colour Matching. in Proceedings - 2015 IEEE International Conference on Multimedia Big Data, BigMM 2015., 7153904, Proceedings - 2015 IEEE International Conference on Multimedia Big Data, BigMM 2015, Institute of Electrical and Electronics Engineers Inc., pp. 300-305, 1st IEEE International Conference on Multimedia Big Data, BigMM 2015, Beijing, China, 20/04/15. https://doi.org/10.1109/BigMM.2015.84

Multi-view Pedestrian Detection Using Statistical Colour Matching. / Ren, Jie; Xu, Ming; Smith, Jeremy S.
Proceedings - 2015 IEEE International Conference on Multimedia Big Data, BigMM 2015. Institute of Electrical and Electronics Engineers Inc., 2015. p. 300-305 7153904 (Proceedings - 2015 IEEE International Conference on Multimedia Big Data, BigMM 2015).

Research output: Chapter in Book or Report/Conference proceeding › Conference Proceeding › peer-review

TY - GEN

T1 - Multi-view Pedestrian Detection Using Statistical Colour Matching

AU - Ren, Jie

AU - Xu, Ming

AU - Smith, Jeremy S.

PY - 2015/7/9

Y1 - 2015/7/9

N2 - To increase the robustness of detection in intelligent video surveillance systems, homography has been widely used to fuse foreground regions projected from multiple camera views to a reference view. The objective of this paper is to detect multiple pedestrians and identify the false-positive detections, which occur due to the foreground intersections of non-corresponding objects, in the top view using occupancy information and colour matching. Multiple homographies are used to detect the head plane and height of each pedestrian. The head locations can be used in the further tracking part. Experimental results show good performance of this method.

AB - To increase the robustness of detection in intelligent video surveillance systems, homography has been widely used to fuse foreground regions projected from multiple camera views to a reference view. The objective of this paper is to detect multiple pedestrians and identify the false-positive detections, which occur due to the foreground intersections of non-corresponding objects, in the top view using occupancy information and colour matching. Multiple homographies are used to detect the head plane and height of each pedestrian. The head locations can be used in the further tracking part. Experimental results show good performance of this method.

KW - compressive sensing

KW - sparse representation

KW - structure set prediction

KW - subspace learning

KW - visual categorization

UR - http://www.scopus.com/inward/record.url?scp=84941218906&partnerID=8YFLogxK

U2 - 10.1109/BigMM.2015.84

DO - 10.1109/BigMM.2015.84

M3 - Conference Proceeding

AN - SCOPUS:84941218906

T3 - Proceedings - 2015 IEEE International Conference on Multimedia Big Data, BigMM 2015

SP - 300

EP - 305

BT - Proceedings - 2015 IEEE International Conference on Multimedia Big Data, BigMM 2015

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 1st IEEE International Conference on Multimedia Big Data, BigMM 2015

Y2 - 20 April 2015 through 22 April 2015

ER -

Ren J, Xu M, Smith JS. Multi-view Pedestrian Detection Using Statistical Colour Matching. In Proceedings - 2015 IEEE International Conference on Multimedia Big Data, BigMM 2015. Institute of Electrical and Electronics Engineers Inc. 2015. p. 300-305. 7153904. (Proceedings - 2015 IEEE International Conference on Multimedia Big Data, BigMM 2015). doi: 10.1109/BigMM.2015.84

Multi-view Pedestrian Detection Using Statistical Colour Matching

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this