Deciphering the crowd: Modeling and identification of pedestrian group motion

Zeynep Yucel, Francesco Zanlungo, Tetsushi Ikeda, Takahiro Miyashita, Norihiro Hagita

Research output: Contribution to journalArticle

16 Citations (Scopus)

Abstract

Associating attributes to pedestrians in a crowd is relevant for various areas like surveillance, customer profiling and service providing. The attributes of interest greatly depend on the application domain and might involve such social relations as friends or family as well as the hierarchy of the group including the leader or subordinates. Nevertheless, the complex social setting inherently complicates this task. We attack this problem by exploiting the small group structures in the crowd. The relations among individuals and their peers within a social group are reliable indicators of social attributes. To that end, this paper identifies social groups based on explicit motion models integrated through a hypothesis testing scheme. We develop two models relating positional and directional relations. A pair of pedestrians is identified as belonging to the same group or not by utilizing the two models in parallel, which defines a compound hypothesis testing scheme. By testing the proposed approach on three datasets with different environmental properties and group characteristics, it is demonstrated that we achieve an identification accuracy of 87% to 99%. The contribution of this study lies in its definition of positional and directional relation models, its description of compound evaluations, and the resolution of ambiguities with our proposed uncertainty measure based on the local and global indicators of group relation.

Original languageEnglish
Pages (from-to)875-897
Number of pages23
JournalSensors (Switzerland)
Volume13
Issue number1
DOIs
Publication statusPublished - Jan 2013
Externally publishedYes

Fingerprint

Social Identification
Group Structure
Uncertainty
Testing
surveillance
Pedestrians
ambiguity
attack
hierarchies
evaluation
Datasets
Sociological Factors

Keywords

  • Motion model
  • Recognition
  • Tracking

ASJC Scopus subject areas

  • Analytical Chemistry
  • Atomic and Molecular Physics, and Optics
  • Biochemistry
  • Electrical and Electronic Engineering

Cite this

Deciphering the crowd : Modeling and identification of pedestrian group motion. / Yucel, Zeynep; Zanlungo, Francesco; Ikeda, Tetsushi; Miyashita, Takahiro; Hagita, Norihiro.

In: Sensors (Switzerland), Vol. 13, No. 1, 01.2013, p. 875-897.

Research output: Contribution to journalArticle

Yucel, Zeynep ; Zanlungo, Francesco ; Ikeda, Tetsushi ; Miyashita, Takahiro ; Hagita, Norihiro. / Deciphering the crowd : Modeling and identification of pedestrian group motion. In: Sensors (Switzerland). 2013 ; Vol. 13, No. 1. pp. 875-897.
@article{e1ab1b81899f4f9cb258ef62b4a14124,
title = "Deciphering the crowd: Modeling and identification of pedestrian group motion",
abstract = "Associating attributes to pedestrians in a crowd is relevant for various areas like surveillance, customer profiling and service providing. The attributes of interest greatly depend on the application domain and might involve such social relations as friends or family as well as the hierarchy of the group including the leader or subordinates. Nevertheless, the complex social setting inherently complicates this task. We attack this problem by exploiting the small group structures in the crowd. The relations among individuals and their peers within a social group are reliable indicators of social attributes. To that end, this paper identifies social groups based on explicit motion models integrated through a hypothesis testing scheme. We develop two models relating positional and directional relations. A pair of pedestrians is identified as belonging to the same group or not by utilizing the two models in parallel, which defines a compound hypothesis testing scheme. By testing the proposed approach on three datasets with different environmental properties and group characteristics, it is demonstrated that we achieve an identification accuracy of 87{\%} to 99{\%}. The contribution of this study lies in its definition of positional and directional relation models, its description of compound evaluations, and the resolution of ambiguities with our proposed uncertainty measure based on the local and global indicators of group relation.",
keywords = "Motion model, Recognition, Tracking",
author = "Zeynep Yucel and Francesco Zanlungo and Tetsushi Ikeda and Takahiro Miyashita and Norihiro Hagita",
year = "2013",
month = "1",
doi = "10.3390/s130100875",
language = "English",
volume = "13",
pages = "875--897",
journal = "Sensors",
issn = "1424-3210",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "1",

}

TY - JOUR

T1 - Deciphering the crowd

T2 - Modeling and identification of pedestrian group motion

AU - Yucel, Zeynep

AU - Zanlungo, Francesco

AU - Ikeda, Tetsushi

AU - Miyashita, Takahiro

AU - Hagita, Norihiro

PY - 2013/1

Y1 - 2013/1

N2 - Associating attributes to pedestrians in a crowd is relevant for various areas like surveillance, customer profiling and service providing. The attributes of interest greatly depend on the application domain and might involve such social relations as friends or family as well as the hierarchy of the group including the leader or subordinates. Nevertheless, the complex social setting inherently complicates this task. We attack this problem by exploiting the small group structures in the crowd. The relations among individuals and their peers within a social group are reliable indicators of social attributes. To that end, this paper identifies social groups based on explicit motion models integrated through a hypothesis testing scheme. We develop two models relating positional and directional relations. A pair of pedestrians is identified as belonging to the same group or not by utilizing the two models in parallel, which defines a compound hypothesis testing scheme. By testing the proposed approach on three datasets with different environmental properties and group characteristics, it is demonstrated that we achieve an identification accuracy of 87% to 99%. The contribution of this study lies in its definition of positional and directional relation models, its description of compound evaluations, and the resolution of ambiguities with our proposed uncertainty measure based on the local and global indicators of group relation.

AB - Associating attributes to pedestrians in a crowd is relevant for various areas like surveillance, customer profiling and service providing. The attributes of interest greatly depend on the application domain and might involve such social relations as friends or family as well as the hierarchy of the group including the leader or subordinates. Nevertheless, the complex social setting inherently complicates this task. We attack this problem by exploiting the small group structures in the crowd. The relations among individuals and their peers within a social group are reliable indicators of social attributes. To that end, this paper identifies social groups based on explicit motion models integrated through a hypothesis testing scheme. We develop two models relating positional and directional relations. A pair of pedestrians is identified as belonging to the same group or not by utilizing the two models in parallel, which defines a compound hypothesis testing scheme. By testing the proposed approach on three datasets with different environmental properties and group characteristics, it is demonstrated that we achieve an identification accuracy of 87% to 99%. The contribution of this study lies in its definition of positional and directional relation models, its description of compound evaluations, and the resolution of ambiguities with our proposed uncertainty measure based on the local and global indicators of group relation.

KW - Motion model

KW - Recognition

KW - Tracking

UR - http://www.scopus.com/inward/record.url?scp=84872914209&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84872914209&partnerID=8YFLogxK

U2 - 10.3390/s130100875

DO - 10.3390/s130100875

M3 - Article

C2 - 23344382

AN - SCOPUS:84872914209

VL - 13

SP - 875

EP - 897

JO - Sensors

JF - Sensors

SN - 1424-3210

IS - 1

ER -