Crowd Space: A predictive crowd analysis technique

Ioannis Karamouzas, Nick Sohre, Ran Hu, Stephen J Guy

Research output: Contribution to journalArticle

Abstract

Over the last two decades there has been a proliferation of methods for simulating crowds of humans. As the number of different methods and their complexity increases, it becomes increasingly unrealistic to expect researchers and users to keep up with all the possible options and trade-offs. We therefore see the need for tools that can facilitate both domain experts and non-expert users of crowd simulation in making high-level decisions about the best simulation methods to use in different scenarios. In this paper, we leverage trajectory data from human crowds and machine learning techniques to learn a manifold which captures representative local navigation scenarios that humans encounter in real life. We show the applicability of this manifold in crowd research, including analyzing trends in simulation accuracy, and creating automated systems to assist in choosing an appropriate simulation method for a given scenario.

Original languageEnglish (US)
Article number186
JournalACM Transactions on Graphics
Volume37
Issue number6
DOIs
StatePublished - Nov 2018

Fingerprint

Learning systems
Navigation
Trajectories
Predictive analytics

Keywords

  • Crowd simulation
  • Entropy
  • Manifold learning
  • Validation

Cite this

Crowd Space : A predictive crowd analysis technique. / Karamouzas, Ioannis; Sohre, Nick; Hu, Ran; Guy, Stephen J.

In: ACM Transactions on Graphics, Vol. 37, No. 6, 186, 11.2018.

Research output: Contribution to journalArticle

Karamouzas, Ioannis ; Sohre, Nick ; Hu, Ran ; Guy, Stephen J. / Crowd Space : A predictive crowd analysis technique. In: ACM Transactions on Graphics. 2018 ; Vol. 37, No. 6.
@article{585097be3bb74d5da4e5be9df9e7599d,
title = "Crowd Space: A predictive crowd analysis technique",
abstract = "Over the last two decades there has been a proliferation of methods for simulating crowds of humans. As the number of different methods and their complexity increases, it becomes increasingly unrealistic to expect researchers and users to keep up with all the possible options and trade-offs. We therefore see the need for tools that can facilitate both domain experts and non-expert users of crowd simulation in making high-level decisions about the best simulation methods to use in different scenarios. In this paper, we leverage trajectory data from human crowds and machine learning techniques to learn a manifold which captures representative local navigation scenarios that humans encounter in real life. We show the applicability of this manifold in crowd research, including analyzing trends in simulation accuracy, and creating automated systems to assist in choosing an appropriate simulation method for a given scenario.",
keywords = "Crowd simulation, Entropy, Manifold learning, Validation",
author = "Ioannis Karamouzas and Nick Sohre and Ran Hu and Guy, {Stephen J}",
year = "2018",
month = "11",
doi = "10.1145/3272127.3275079",
language = "English (US)",
volume = "37",
journal = "ACM Transactions on Computer Systems",
issn = "0730-0301",
publisher = "Association for Computing Machinery (ACM)",
number = "6",

}

TY - JOUR

T1 - Crowd Space

T2 - A predictive crowd analysis technique

AU - Karamouzas, Ioannis

AU - Sohre, Nick

AU - Hu, Ran

AU - Guy, Stephen J

PY - 2018/11

Y1 - 2018/11

N2 - Over the last two decades there has been a proliferation of methods for simulating crowds of humans. As the number of different methods and their complexity increases, it becomes increasingly unrealistic to expect researchers and users to keep up with all the possible options and trade-offs. We therefore see the need for tools that can facilitate both domain experts and non-expert users of crowd simulation in making high-level decisions about the best simulation methods to use in different scenarios. In this paper, we leverage trajectory data from human crowds and machine learning techniques to learn a manifold which captures representative local navigation scenarios that humans encounter in real life. We show the applicability of this manifold in crowd research, including analyzing trends in simulation accuracy, and creating automated systems to assist in choosing an appropriate simulation method for a given scenario.

AB - Over the last two decades there has been a proliferation of methods for simulating crowds of humans. As the number of different methods and their complexity increases, it becomes increasingly unrealistic to expect researchers and users to keep up with all the possible options and trade-offs. We therefore see the need for tools that can facilitate both domain experts and non-expert users of crowd simulation in making high-level decisions about the best simulation methods to use in different scenarios. In this paper, we leverage trajectory data from human crowds and machine learning techniques to learn a manifold which captures representative local navigation scenarios that humans encounter in real life. We show the applicability of this manifold in crowd research, including analyzing trends in simulation accuracy, and creating automated systems to assist in choosing an appropriate simulation method for a given scenario.

KW - Crowd simulation

KW - Entropy

KW - Manifold learning

KW - Validation

UR - http://www.scopus.com/inward/record.url?scp=85064835225&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85064835225&partnerID=8YFLogxK

U2 - 10.1145/3272127.3275079

DO - 10.1145/3272127.3275079

M3 - Article

AN - SCOPUS:85064835225

VL - 37

JO - ACM Transactions on Computer Systems

JF - ACM Transactions on Computer Systems

SN - 0730-0301

IS - 6

M1 - 186

ER -