Surveillance camera autocalibration based on pedestrian height distributions

Jingchen Liu, Robert Collins, Yanxi Liu

Research output: Contribution to conferencePaper

31 Citations (Scopus)

Abstract

We propose a new framework for automatic surveillance camera calibration by observing videos of pedestrians walking through the scene. Unlike existing methods that require accurate pedestrian detection and tracking, our method takes noisy foreground masks as input and automatically estimates the necessary intrinsic and extrinsic camera parameters using prior knowledge about the distribution of relative human heights. Our algorithm is computationally efficient enough for online parameter estimation. Experimental results on both synthetic and real data show the robustness of our method to camera pose and noisy foreground detections.

Original languageEnglish (US)
DOIs
StatePublished - Jan 1 2011
Event2011 22nd British Machine Vision Conference, BMVC 2011 - Dundee, United Kingdom
Duration: Aug 29 2011Sep 2 2011

Other

Other2011 22nd British Machine Vision Conference, BMVC 2011
CountryUnited Kingdom
CityDundee
Period8/29/119/2/11

Fingerprint

Cameras
Parameter estimation
Masks
Calibration

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition

Cite this

Liu, J., Collins, R., & Liu, Y. (2011). Surveillance camera autocalibration based on pedestrian height distributions. Paper presented at 2011 22nd British Machine Vision Conference, BMVC 2011, Dundee, United Kingdom. https://doi.org/10.5244/C.25.117
Liu, Jingchen ; Collins, Robert ; Liu, Yanxi. / Surveillance camera autocalibration based on pedestrian height distributions. Paper presented at 2011 22nd British Machine Vision Conference, BMVC 2011, Dundee, United Kingdom.
@conference{5aa829aff6204c8d8e24ab6cf42138c0,
title = "Surveillance camera autocalibration based on pedestrian height distributions",
abstract = "We propose a new framework for automatic surveillance camera calibration by observing videos of pedestrians walking through the scene. Unlike existing methods that require accurate pedestrian detection and tracking, our method takes noisy foreground masks as input and automatically estimates the necessary intrinsic and extrinsic camera parameters using prior knowledge about the distribution of relative human heights. Our algorithm is computationally efficient enough for online parameter estimation. Experimental results on both synthetic and real data show the robustness of our method to camera pose and noisy foreground detections.",
author = "Jingchen Liu and Robert Collins and Yanxi Liu",
year = "2011",
month = "1",
day = "1",
doi = "10.5244/C.25.117",
language = "English (US)",
note = "2011 22nd British Machine Vision Conference, BMVC 2011 ; Conference date: 29-08-2011 Through 02-09-2011",

}

Liu, J, Collins, R & Liu, Y 2011, 'Surveillance camera autocalibration based on pedestrian height distributions' Paper presented at 2011 22nd British Machine Vision Conference, BMVC 2011, Dundee, United Kingdom, 8/29/11 - 9/2/11, . https://doi.org/10.5244/C.25.117

Surveillance camera autocalibration based on pedestrian height distributions. / Liu, Jingchen; Collins, Robert; Liu, Yanxi.

2011. Paper presented at 2011 22nd British Machine Vision Conference, BMVC 2011, Dundee, United Kingdom.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Surveillance camera autocalibration based on pedestrian height distributions

AU - Liu, Jingchen

AU - Collins, Robert

AU - Liu, Yanxi

PY - 2011/1/1

Y1 - 2011/1/1

N2 - We propose a new framework for automatic surveillance camera calibration by observing videos of pedestrians walking through the scene. Unlike existing methods that require accurate pedestrian detection and tracking, our method takes noisy foreground masks as input and automatically estimates the necessary intrinsic and extrinsic camera parameters using prior knowledge about the distribution of relative human heights. Our algorithm is computationally efficient enough for online parameter estimation. Experimental results on both synthetic and real data show the robustness of our method to camera pose and noisy foreground detections.

AB - We propose a new framework for automatic surveillance camera calibration by observing videos of pedestrians walking through the scene. Unlike existing methods that require accurate pedestrian detection and tracking, our method takes noisy foreground masks as input and automatically estimates the necessary intrinsic and extrinsic camera parameters using prior knowledge about the distribution of relative human heights. Our algorithm is computationally efficient enough for online parameter estimation. Experimental results on both synthetic and real data show the robustness of our method to camera pose and noisy foreground detections.

UR - http://www.scopus.com/inward/record.url?scp=84898446159&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84898446159&partnerID=8YFLogxK

U2 - 10.5244/C.25.117

DO - 10.5244/C.25.117

M3 - Paper

AN - SCOPUS:84898446159

ER -

Liu J, Collins R, Liu Y. Surveillance camera autocalibration based on pedestrian height distributions. 2011. Paper presented at 2011 22nd British Machine Vision Conference, BMVC 2011, Dundee, United Kingdom. https://doi.org/10.5244/C.25.117