Simple real-time human detection using a single correlation filter

David S. Bolme; Yui Man Lui; Bruce A. Draper; J. Ross Beveridge

doi:10.1109/PETS-WINTER.2009.5399555

Simple real-time human detection using a single correlation filter

David S. Bolme, Yui Man Lui, Bruce A. Draper, J. Ross Beveridge

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

35 Scopus citations

Abstract

This paper presents an extremely simple human detection algorithm based on correlating edge magnitude images with a filter. The key is the technology used to train the filter: Average of Synthetic Exact Filters (ASEF). The ASEF based detector can process images at over 25 frames per second and achieves a 94.5% detection rate with less than one false detection per frame for sparse crowds. Filter training is also fast, taking only 12 seconds to train the detector on 32 manually annotated images. Evaluation is performed on the PETS 2009 dataset and results are compared to the OpenCV cascade classifier and a state-of-the-art deformable parts based person detector.

Original language	English
Title of host publication	Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009
DOIs	https://doi.org/10.1109/PETS-WINTER.2009.5399555
State	Published - 2009
Externally published	Yes
Event	12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009 - Snowbird, UT, United States Duration: Dec 7 2009 → Dec 9 2009

Publication series

Name	Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009

Conference

Conference	12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009
Country/Territory	United States
City	Snowbird, UT
Period	12/7/09 → 12/9/09

Access to Document

10.1109/PETS-WINTER.2009.5399555

Cite this

Bolme, D. S., Lui, Y. M., Draper, B. A., & Beveridge, J. R. (2009). Simple real-time human detection using a single correlation filter. In Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009 Article 5399555 (Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009). https://doi.org/10.1109/PETS-WINTER.2009.5399555

Bolme, David S. ; Lui, Yui Man ; Draper, Bruce A. et al. / Simple real-time human detection using a single correlation filter. Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009. 2009. (Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009).

@inproceedings{2841551a6a134b058cf7e10dd9a5808e,

title = "Simple real-time human detection using a single correlation filter",

abstract = "This paper presents an extremely simple human detection algorithm based on correlating edge magnitude images with a filter. The key is the technology used to train the filter: Average of Synthetic Exact Filters (ASEF). The ASEF based detector can process images at over 25 frames per second and achieves a 94.5% detection rate with less than one false detection per frame for sparse crowds. Filter training is also fast, taking only 12 seconds to train the detector on 32 manually annotated images. Evaluation is performed on the PETS 2009 dataset and results are compared to the OpenCV cascade classifier and a state-of-the-art deformable parts based person detector.",

author = "Bolme, {David S.} and Lui, {Yui Man} and Draper, {Bruce A.} and Beveridge, {J. Ross}",

year = "2009",

doi = "10.1109/PETS-WINTER.2009.5399555",

language = "English",

isbn = "9781424455034",

series = "Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009",

booktitle = "Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009",

note = "12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009 ; Conference date: 07-12-2009 Through 09-12-2009",

}

Bolme, DS, Lui, YM, Draper, BA & Beveridge, JR 2009, Simple real-time human detection using a single correlation filter. in Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009., 5399555, Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009, 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009, Snowbird, UT, United States, 12/7/09. https://doi.org/10.1109/PETS-WINTER.2009.5399555

Simple real-time human detection using a single correlation filter. / Bolme, David S.; Lui, Yui Man; Draper, Bruce A. et al.
Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009. 2009. 5399555 (Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009).

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

TY - GEN

T1 - Simple real-time human detection using a single correlation filter

AU - Bolme, David S.

AU - Lui, Yui Man

AU - Draper, Bruce A.

AU - Beveridge, J. Ross

PY - 2009

Y1 - 2009

N2 - This paper presents an extremely simple human detection algorithm based on correlating edge magnitude images with a filter. The key is the technology used to train the filter: Average of Synthetic Exact Filters (ASEF). The ASEF based detector can process images at over 25 frames per second and achieves a 94.5% detection rate with less than one false detection per frame for sparse crowds. Filter training is also fast, taking only 12 seconds to train the detector on 32 manually annotated images. Evaluation is performed on the PETS 2009 dataset and results are compared to the OpenCV cascade classifier and a state-of-the-art deformable parts based person detector.

AB - This paper presents an extremely simple human detection algorithm based on correlating edge magnitude images with a filter. The key is the technology used to train the filter: Average of Synthetic Exact Filters (ASEF). The ASEF based detector can process images at over 25 frames per second and achieves a 94.5% detection rate with less than one false detection per frame for sparse crowds. Filter training is also fast, taking only 12 seconds to train the detector on 32 manually annotated images. Evaluation is performed on the PETS 2009 dataset and results are compared to the OpenCV cascade classifier and a state-of-the-art deformable parts based person detector.

UR - http://www.scopus.com/inward/record.url?scp=77950622491&partnerID=8YFLogxK

U2 - 10.1109/PETS-WINTER.2009.5399555

DO - 10.1109/PETS-WINTER.2009.5399555

M3 - Conference contribution

AN - SCOPUS:77950622491

SN - 9781424455034

T3 - Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009

BT - Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009

T2 - 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009

Y2 - 7 December 2009 through 9 December 2009

ER -

Bolme DS, Lui YM, Draper BA, Beveridge JR. Simple real-time human detection using a single correlation filter. In Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009. 2009. 5399555. (Proceedings of the 12th IEEE International Workshop on Performance Evaluation of Tracking and Surveillance, PETS-Winter 2009). doi: 10.1109/PETS-WINTER.2009.5399555

Simple real-time human detection using a single correlation filter

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this