Human detection using motion and appearance based feature

Shaopeng Tang, Satoshi Goto

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

3 Citations (Scopus)

Abstract

An approach to detect moving and standing human from video is proposed in this paper. Human detection from videos is a difficult problem because of motion of human, camera and background. In order to detect moving human, the dense optical flow is calculated by two consecutive frames, to represent the motion of human. Motion based feature is extracted from optical flow field. It not only represents the global motion caused by the boundary of human body, but also contains local motion caused by the limbs. Motion based feature is combined with histogram of template feature, which is designed to detect standing human, as final feature for detection. Experiment on CAS dataset shows that this feature has more discriminative ability than other motion based feature. Besides, this feature is easier for hardware acceleration, which makes it suitable for real time application.

Original languageEnglish
Title of host publicationICICS 2009 - Conference Proceedings of the 7th International Conference on Information, Communications and Signal Processing
DOIs
Publication statusPublished - 2009
Event7th International Conference on Information, Communications and Signal Processing, ICICS 2009 - Macau Fisherman's Wharf
Duration: 2009 Dec 82009 Dec 10

Other

Other7th International Conference on Information, Communications and Signal Processing, ICICS 2009
CityMacau Fisherman's Wharf
Period09/12/809/12/10

Fingerprint

Optical flows
Flow fields
Cameras
Hardware
Experiments

Keywords

  • Histogram of template
  • Human detection
  • Optical flow

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Information Systems
  • Signal Processing

Cite this

Tang, S., & Goto, S. (2009). Human detection using motion and appearance based feature. In ICICS 2009 - Conference Proceedings of the 7th International Conference on Information, Communications and Signal Processing [5397675] https://doi.org/10.1109/ICICS.2009.5397675

Human detection using motion and appearance based feature. / Tang, Shaopeng; Goto, Satoshi.

ICICS 2009 - Conference Proceedings of the 7th International Conference on Information, Communications and Signal Processing. 2009. 5397675.

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

Tang, S & Goto, S 2009, Human detection using motion and appearance based feature. in ICICS 2009 - Conference Proceedings of the 7th International Conference on Information, Communications and Signal Processing., 5397675, 7th International Conference on Information, Communications and Signal Processing, ICICS 2009, Macau Fisherman's Wharf, 09/12/8. https://doi.org/10.1109/ICICS.2009.5397675
Tang S, Goto S. Human detection using motion and appearance based feature. In ICICS 2009 - Conference Proceedings of the 7th International Conference on Information, Communications and Signal Processing. 2009. 5397675 https://doi.org/10.1109/ICICS.2009.5397675
Tang, Shaopeng ; Goto, Satoshi. / Human detection using motion and appearance based feature. ICICS 2009 - Conference Proceedings of the 7th International Conference on Information, Communications and Signal Processing. 2009.
@inproceedings{54359b770b2f4236b692e2b47896ad3e,
title = "Human detection using motion and appearance based feature",
abstract = "An approach to detect moving and standing human from video is proposed in this paper. Human detection from videos is a difficult problem because of motion of human, camera and background. In order to detect moving human, the dense optical flow is calculated by two consecutive frames, to represent the motion of human. Motion based feature is extracted from optical flow field. It not only represents the global motion caused by the boundary of human body, but also contains local motion caused by the limbs. Motion based feature is combined with histogram of template feature, which is designed to detect standing human, as final feature for detection. Experiment on CAS dataset shows that this feature has more discriminative ability than other motion based feature. Besides, this feature is easier for hardware acceleration, which makes it suitable for real time application.",
keywords = "Histogram of template, Human detection, Optical flow",
author = "Shaopeng Tang and Satoshi Goto",
year = "2009",
doi = "10.1109/ICICS.2009.5397675",
language = "English",
isbn = "9781424446575",
booktitle = "ICICS 2009 - Conference Proceedings of the 7th International Conference on Information, Communications and Signal Processing",

}

TY - GEN

T1 - Human detection using motion and appearance based feature

AU - Tang, Shaopeng

AU - Goto, Satoshi

PY - 2009

Y1 - 2009

N2 - An approach to detect moving and standing human from video is proposed in this paper. Human detection from videos is a difficult problem because of motion of human, camera and background. In order to detect moving human, the dense optical flow is calculated by two consecutive frames, to represent the motion of human. Motion based feature is extracted from optical flow field. It not only represents the global motion caused by the boundary of human body, but also contains local motion caused by the limbs. Motion based feature is combined with histogram of template feature, which is designed to detect standing human, as final feature for detection. Experiment on CAS dataset shows that this feature has more discriminative ability than other motion based feature. Besides, this feature is easier for hardware acceleration, which makes it suitable for real time application.

AB - An approach to detect moving and standing human from video is proposed in this paper. Human detection from videos is a difficult problem because of motion of human, camera and background. In order to detect moving human, the dense optical flow is calculated by two consecutive frames, to represent the motion of human. Motion based feature is extracted from optical flow field. It not only represents the global motion caused by the boundary of human body, but also contains local motion caused by the limbs. Motion based feature is combined with histogram of template feature, which is designed to detect standing human, as final feature for detection. Experiment on CAS dataset shows that this feature has more discriminative ability than other motion based feature. Besides, this feature is easier for hardware acceleration, which makes it suitable for real time application.

KW - Histogram of template

KW - Human detection

KW - Optical flow

UR - http://www.scopus.com/inward/record.url?scp=77949606451&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77949606451&partnerID=8YFLogxK

U2 - 10.1109/ICICS.2009.5397675

DO - 10.1109/ICICS.2009.5397675

M3 - Conference contribution

AN - SCOPUS:77949606451

SN - 9781424446575

BT - ICICS 2009 - Conference Proceedings of the 7th International Conference on Information, Communications and Signal Processing

ER -