Tracking People in Dense Crowds Using Supervoxels

Shota Takayama, Teppei Suzuki, Yoshimitsu Aoki, Sho Isobe, Makoto Masuda

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

Abstract

The demand for people tracking in dense crowds is increasing, but it is a challenging problem in the computer vision field. 'Crowd tracking' is extremely difficult because of hard occlusions, various motions and posture changes. In particular, we need to handle occlusions for more robust tracking. This paper discusses robust crowd tracking based on a combination of supervoxels and optical flow tracking. The SLIC based supervoxel algorithm adaptively estimates the boundary between a person and a background. Therefore, the combination of supervoxels and optical flow tracking becomes a highly reliable approach for crowd tracking. In tracking experiments, high performance is achieved for the UCF crowd dataset.

Original languageEnglish
Title of host publicationProceedings - 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages532-537
Number of pages6
ISBN (Electronic)9781509056989
DOIs
Publication statusPublished - 2017 Apr 21
Event12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016 - Naples, Italy
Duration: 2016 Nov 282016 Dec 1

Other

Other12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016
CountryItaly
CityNaples
Period16/11/2816/12/1

Fingerprint

Optical flows
Posture
Computer vision
Experiments
Datasets

Keywords

  • dense crowd
  • optical flow
  • People tracking
  • supervoxel

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Radiology Nuclear Medicine and imaging
  • Computer Networks and Communications
  • Signal Processing

Cite this

Takayama, S., Suzuki, T., Aoki, Y., Isobe, S., & Masuda, M. (2017). Tracking People in Dense Crowds Using Supervoxels. In Proceedings - 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016 (pp. 532-537). [7907517] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/SITIS.2016.90

Tracking People in Dense Crowds Using Supervoxels. / Takayama, Shota; Suzuki, Teppei; Aoki, Yoshimitsu; Isobe, Sho; Masuda, Makoto.

Proceedings - 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016. Institute of Electrical and Electronics Engineers Inc., 2017. p. 532-537 7907517.

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

Takayama, S, Suzuki, T, Aoki, Y, Isobe, S & Masuda, M 2017, Tracking People in Dense Crowds Using Supervoxels. in Proceedings - 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016., 7907517, Institute of Electrical and Electronics Engineers Inc., pp. 532-537, 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016, Naples, Italy, 16/11/28. https://doi.org/10.1109/SITIS.2016.90
Takayama S, Suzuki T, Aoki Y, Isobe S, Masuda M. Tracking People in Dense Crowds Using Supervoxels. In Proceedings - 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016. Institute of Electrical and Electronics Engineers Inc. 2017. p. 532-537. 7907517 https://doi.org/10.1109/SITIS.2016.90
Takayama, Shota ; Suzuki, Teppei ; Aoki, Yoshimitsu ; Isobe, Sho ; Masuda, Makoto. / Tracking People in Dense Crowds Using Supervoxels. Proceedings - 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016. Institute of Electrical and Electronics Engineers Inc., 2017. pp. 532-537
@inproceedings{3503a8cc37ec48599bd87fef522f4b99,
title = "Tracking People in Dense Crowds Using Supervoxels",
abstract = "The demand for people tracking in dense crowds is increasing, but it is a challenging problem in the computer vision field. 'Crowd tracking' is extremely difficult because of hard occlusions, various motions and posture changes. In particular, we need to handle occlusions for more robust tracking. This paper discusses robust crowd tracking based on a combination of supervoxels and optical flow tracking. The SLIC based supervoxel algorithm adaptively estimates the boundary between a person and a background. Therefore, the combination of supervoxels and optical flow tracking becomes a highly reliable approach for crowd tracking. In tracking experiments, high performance is achieved for the UCF crowd dataset.",
keywords = "dense crowd, optical flow, People tracking, supervoxel",
author = "Shota Takayama and Teppei Suzuki and Yoshimitsu Aoki and Sho Isobe and Makoto Masuda",
year = "2017",
month = "4",
day = "21",
doi = "10.1109/SITIS.2016.90",
language = "English",
pages = "532--537",
booktitle = "Proceedings - 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Tracking People in Dense Crowds Using Supervoxels

AU - Takayama, Shota

AU - Suzuki, Teppei

AU - Aoki, Yoshimitsu

AU - Isobe, Sho

AU - Masuda, Makoto

PY - 2017/4/21

Y1 - 2017/4/21

N2 - The demand for people tracking in dense crowds is increasing, but it is a challenging problem in the computer vision field. 'Crowd tracking' is extremely difficult because of hard occlusions, various motions and posture changes. In particular, we need to handle occlusions for more robust tracking. This paper discusses robust crowd tracking based on a combination of supervoxels and optical flow tracking. The SLIC based supervoxel algorithm adaptively estimates the boundary between a person and a background. Therefore, the combination of supervoxels and optical flow tracking becomes a highly reliable approach for crowd tracking. In tracking experiments, high performance is achieved for the UCF crowd dataset.

AB - The demand for people tracking in dense crowds is increasing, but it is a challenging problem in the computer vision field. 'Crowd tracking' is extremely difficult because of hard occlusions, various motions and posture changes. In particular, we need to handle occlusions for more robust tracking. This paper discusses robust crowd tracking based on a combination of supervoxels and optical flow tracking. The SLIC based supervoxel algorithm adaptively estimates the boundary between a person and a background. Therefore, the combination of supervoxels and optical flow tracking becomes a highly reliable approach for crowd tracking. In tracking experiments, high performance is achieved for the UCF crowd dataset.

KW - dense crowd

KW - optical flow

KW - People tracking

KW - supervoxel

UR - http://www.scopus.com/inward/record.url?scp=85019175093&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85019175093&partnerID=8YFLogxK

U2 - 10.1109/SITIS.2016.90

DO - 10.1109/SITIS.2016.90

M3 - Conference contribution

AN - SCOPUS:85019175093

SP - 532

EP - 537

BT - Proceedings - 12th International Conference on Signal Image Technology and Internet-Based Systems, SITIS 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -