Marked point processes for crowd counting

Weina Ge, Robert Collins

Research output: Chapter in Book/Report/Conference proceedingConference contribution

131 Citations (Scopus)

Abstract

A Bayesian marked point process (MPP) model is developed to detect and count people in crowded scenes. The model couples a spatial stochastic process governing number and placement of individuals with a conditional mark process for selecting body shape. We automatically learn the mark (shape) process from training video by estimating a mixture of Bernoulli shape prototypes along with an extrinsic shape distribution describing the orientation and scaling of these shapes for any given image location. The reversible jump Markov Chain Monte Carlo framework is used to efficiently search for the maximum a posteriori configuration of shapes, leading to an estimate of the count, location and pose of each person in the scene. Quantitative results of crowd counting are presented for two publiclyavailable datasets with known ground truth.

Original languageEnglish (US)
Title of host publication2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009
PublisherIEEE Computer Society
Pages2913-2920
Number of pages8
Volume2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
ISBN (Print)9781424439935
DOIs
StatePublished - Jan 1 2009
Event2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Miami, FL, United States
Duration: Jun 20 2009Jun 25 2009

Other

Other2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
CountryUnited States
CityMiami, FL
Period6/20/096/25/09

Fingerprint

Random processes
Markov processes

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition
  • Biomedical Engineering

Cite this

Ge, W., & Collins, R. (2009). Marked point processes for crowd counting. In 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009 (Vol. 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 2913-2920). [5206621] IEEE Computer Society. https://doi.org/10.1109/CVPRW.2009.5206621
Ge, Weina ; Collins, Robert. / Marked point processes for crowd counting. 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009. Vol. 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition IEEE Computer Society, 2009. pp. 2913-2920
@inproceedings{abbd209e7cbd4f0eaca7a261dc381608,
title = "Marked point processes for crowd counting",
abstract = "A Bayesian marked point process (MPP) model is developed to detect and count people in crowded scenes. The model couples a spatial stochastic process governing number and placement of individuals with a conditional mark process for selecting body shape. We automatically learn the mark (shape) process from training video by estimating a mixture of Bernoulli shape prototypes along with an extrinsic shape distribution describing the orientation and scaling of these shapes for any given image location. The reversible jump Markov Chain Monte Carlo framework is used to efficiently search for the maximum a posteriori configuration of shapes, leading to an estimate of the count, location and pose of each person in the scene. Quantitative results of crowd counting are presented for two publiclyavailable datasets with known ground truth.",
author = "Weina Ge and Robert Collins",
year = "2009",
month = "1",
day = "1",
doi = "10.1109/CVPRW.2009.5206621",
language = "English (US)",
isbn = "9781424439935",
volume = "2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition",
pages = "2913--2920",
booktitle = "2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009",
publisher = "IEEE Computer Society",
address = "United States",

}

Ge, W & Collins, R 2009, Marked point processes for crowd counting. in 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009. vol. 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 5206621, IEEE Computer Society, pp. 2913-2920, 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Miami, FL, United States, 6/20/09. https://doi.org/10.1109/CVPRW.2009.5206621

Marked point processes for crowd counting. / Ge, Weina; Collins, Robert.

2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009. Vol. 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition IEEE Computer Society, 2009. p. 2913-2920 5206621.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Marked point processes for crowd counting

AU - Ge, Weina

AU - Collins, Robert

PY - 2009/1/1

Y1 - 2009/1/1

N2 - A Bayesian marked point process (MPP) model is developed to detect and count people in crowded scenes. The model couples a spatial stochastic process governing number and placement of individuals with a conditional mark process for selecting body shape. We automatically learn the mark (shape) process from training video by estimating a mixture of Bernoulli shape prototypes along with an extrinsic shape distribution describing the orientation and scaling of these shapes for any given image location. The reversible jump Markov Chain Monte Carlo framework is used to efficiently search for the maximum a posteriori configuration of shapes, leading to an estimate of the count, location and pose of each person in the scene. Quantitative results of crowd counting are presented for two publiclyavailable datasets with known ground truth.

AB - A Bayesian marked point process (MPP) model is developed to detect and count people in crowded scenes. The model couples a spatial stochastic process governing number and placement of individuals with a conditional mark process for selecting body shape. We automatically learn the mark (shape) process from training video by estimating a mixture of Bernoulli shape prototypes along with an extrinsic shape distribution describing the orientation and scaling of these shapes for any given image location. The reversible jump Markov Chain Monte Carlo framework is used to efficiently search for the maximum a posteriori configuration of shapes, leading to an estimate of the count, location and pose of each person in the scene. Quantitative results of crowd counting are presented for two publiclyavailable datasets with known ground truth.

UR - http://www.scopus.com/inward/record.url?scp=70450161263&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70450161263&partnerID=8YFLogxK

U2 - 10.1109/CVPRW.2009.5206621

DO - 10.1109/CVPRW.2009.5206621

M3 - Conference contribution

AN - SCOPUS:70450161263

SN - 9781424439935

VL - 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition

SP - 2913

EP - 2920

BT - 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009

PB - IEEE Computer Society

ER -

Ge W, Collins R. Marked point processes for crowd counting. In 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPR Workshops 2009. Vol. 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE Computer Society. 2009. p. 2913-2920. 5206621 https://doi.org/10.1109/CVPRW.2009.5206621