Fast human detection using a cascade of histograms of oriented gradients

Qiang Zhu, Shai Avidan, Mei-Chen Yeh, Kwang Ting Cheng

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

1097 Citations (Scopus)

Abstract

We integrate the cascade-of-rejectors approach with the Histograms of Oriented Gradients (HoG) features to achieve a fast and accurate human detection system. The features used in our system are HoGs of variable-size blocks that capture salient features of humans automatically. Using AdaBoost for feature selection, we identify the appropriate set of blocks, from a large set of possible blocks. In our system, we use the integral image representation and a rejection cascade which significantly speed up the computation. For a 320 × 250 image, the system can process 5 to 30 frames per second depending on the density in which we scan the image, while maintaining an accuracy level similar to existing methods.

Original languageEnglish
Title of host publicationProceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006
Pages1491-1498
Number of pages8
DOIs
Publication statusPublished - 2006 Dec 22
Event2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006 - New York, NY, United States
Duration: 2006 Jun 172006 Jun 22

Publication series

NameProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Volume2
ISSN (Print)1063-6919

Other

Other2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006
CountryUnited States
CityNew York, NY
Period06/6/1706/6/22

Fingerprint

Adaptive boosting
Feature extraction

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Zhu, Q., Avidan, S., Yeh, M-C., & Cheng, K. T. (2006). Fast human detection using a cascade of histograms of oriented gradients. In Proceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006 (pp. 1491-1498). [1640933] (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition; Vol. 2). https://doi.org/10.1109/CVPR.2006.119

Fast human detection using a cascade of histograms of oriented gradients. / Zhu, Qiang; Avidan, Shai; Yeh, Mei-Chen; Cheng, Kwang Ting.

Proceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006. 2006. p. 1491-1498 1640933 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition; Vol. 2).

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

Zhu, Q, Avidan, S, Yeh, M-C & Cheng, KT 2006, Fast human detection using a cascade of histograms of oriented gradients. in Proceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006., 1640933, Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, vol. 2, pp. 1491-1498, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006, New York, NY, United States, 06/6/17. https://doi.org/10.1109/CVPR.2006.119
Zhu Q, Avidan S, Yeh M-C, Cheng KT. Fast human detection using a cascade of histograms of oriented gradients. In Proceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006. 2006. p. 1491-1498. 1640933. (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition). https://doi.org/10.1109/CVPR.2006.119
Zhu, Qiang ; Avidan, Shai ; Yeh, Mei-Chen ; Cheng, Kwang Ting. / Fast human detection using a cascade of histograms of oriented gradients. Proceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006. 2006. pp. 1491-1498 (Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition).
@inproceedings{ffaf2ce0bad840f3b77135a0be9cea7d,
title = "Fast human detection using a cascade of histograms of oriented gradients",
abstract = "We integrate the cascade-of-rejectors approach with the Histograms of Oriented Gradients (HoG) features to achieve a fast and accurate human detection system. The features used in our system are HoGs of variable-size blocks that capture salient features of humans automatically. Using AdaBoost for feature selection, we identify the appropriate set of blocks, from a large set of possible blocks. In our system, we use the integral image representation and a rejection cascade which significantly speed up the computation. For a 320 × 250 image, the system can process 5 to 30 frames per second depending on the density in which we scan the image, while maintaining an accuracy level similar to existing methods.",
author = "Qiang Zhu and Shai Avidan and Mei-Chen Yeh and Cheng, {Kwang Ting}",
year = "2006",
month = "12",
day = "22",
doi = "10.1109/CVPR.2006.119",
language = "English",
isbn = "0769525970",
series = "Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition",
pages = "1491--1498",
booktitle = "Proceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006",

}

TY - GEN

T1 - Fast human detection using a cascade of histograms of oriented gradients

AU - Zhu, Qiang

AU - Avidan, Shai

AU - Yeh, Mei-Chen

AU - Cheng, Kwang Ting

PY - 2006/12/22

Y1 - 2006/12/22

N2 - We integrate the cascade-of-rejectors approach with the Histograms of Oriented Gradients (HoG) features to achieve a fast and accurate human detection system. The features used in our system are HoGs of variable-size blocks that capture salient features of humans automatically. Using AdaBoost for feature selection, we identify the appropriate set of blocks, from a large set of possible blocks. In our system, we use the integral image representation and a rejection cascade which significantly speed up the computation. For a 320 × 250 image, the system can process 5 to 30 frames per second depending on the density in which we scan the image, while maintaining an accuracy level similar to existing methods.

AB - We integrate the cascade-of-rejectors approach with the Histograms of Oriented Gradients (HoG) features to achieve a fast and accurate human detection system. The features used in our system are HoGs of variable-size blocks that capture salient features of humans automatically. Using AdaBoost for feature selection, we identify the appropriate set of blocks, from a large set of possible blocks. In our system, we use the integral image representation and a rejection cascade which significantly speed up the computation. For a 320 × 250 image, the system can process 5 to 30 frames per second depending on the density in which we scan the image, while maintaining an accuracy level similar to existing methods.

UR - http://www.scopus.com/inward/record.url?scp=33845597667&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33845597667&partnerID=8YFLogxK

U2 - 10.1109/CVPR.2006.119

DO - 10.1109/CVPR.2006.119

M3 - Conference contribution

AN - SCOPUS:33845597667

SN - 0769525970

SN - 9780769525976

T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition

SP - 1491

EP - 1498

BT - Proceedings - 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, CVPR 2006

ER -