A network of dynamic probabilistic models for human interaction analysis

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

We propose a novel method of analyzing human interactions based on the walking trajectories of human subjects, which provide elementary and necessary components for understanding and interpretation of complex human interactions in visual surveillance tasks. Our principal assumption is that an interaction episode is composed of meaningful small unit interactions, which we call sub-interactions. We model each sub-interaction by a dynamic probabilistic model and propose a modified factorial hidden Markov model (HMM) with factored observations. The complete interaction is represented with a network of dynamic probabilistic models (DPMs) by an ordered concatenation of sub-interaction models. The rationale for this approach is that it is more effective in utilizing common components, i.e., sub-interaction models, to describe complex interaction patterns. By assembling these sub-interaction models in a network, possibly with a mixture of different types of DPMs, such as standard HMMs, variants of HMMs, dynamic Bayesian networks, and so on, we can design a robust model for the analysis of human interactions. We show the feasibility and effectiveness of the proposed method by analyzing the structure of network of DPMs and its success on four different databases: a self-collected dataset, Tsinghua University's dataset, the public domain CAVIAR dataset, and the Edinburgh Informatics Forum Pedestrian dataset.

Original languageEnglish
Article number5740319
Pages (from-to)932-945
Number of pages14
JournalIEEE Transactions on Circuits and Systems for Video Technology
Volume21
Issue number7
DOIs
Publication statusPublished - 2011 Jul 1

Fingerprint

Bayesian networks
Hidden Markov models
Trajectories
Statistical Models

Keywords

  • Dynamic Bayesian network
  • human interaction analysis
  • network of dynamic probabilistic models
  • sub-interactions
  • video surveillance

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Media Technology

Cite this

A network of dynamic probabilistic models for human interaction analysis. / Suk, Heung-Il; Jain, Anil K.; Lee, Seong Whan.

In: IEEE Transactions on Circuits and Systems for Video Technology, Vol. 21, No. 7, 5740319, 01.07.2011, p. 932-945.

Research output: Contribution to journalArticle

@article{c6623d19fa2a4d4fa4bbe516e91eef35,
title = "A network of dynamic probabilistic models for human interaction analysis",
abstract = "We propose a novel method of analyzing human interactions based on the walking trajectories of human subjects, which provide elementary and necessary components for understanding and interpretation of complex human interactions in visual surveillance tasks. Our principal assumption is that an interaction episode is composed of meaningful small unit interactions, which we call sub-interactions. We model each sub-interaction by a dynamic probabilistic model and propose a modified factorial hidden Markov model (HMM) with factored observations. The complete interaction is represented with a network of dynamic probabilistic models (DPMs) by an ordered concatenation of sub-interaction models. The rationale for this approach is that it is more effective in utilizing common components, i.e., sub-interaction models, to describe complex interaction patterns. By assembling these sub-interaction models in a network, possibly with a mixture of different types of DPMs, such as standard HMMs, variants of HMMs, dynamic Bayesian networks, and so on, we can design a robust model for the analysis of human interactions. We show the feasibility and effectiveness of the proposed method by analyzing the structure of network of DPMs and its success on four different databases: a self-collected dataset, Tsinghua University's dataset, the public domain CAVIAR dataset, and the Edinburgh Informatics Forum Pedestrian dataset.",
keywords = "Dynamic Bayesian network, human interaction analysis, network of dynamic probabilistic models, sub-interactions, video surveillance",
author = "Heung-Il Suk and Jain, {Anil K.} and Lee, {Seong Whan}",
year = "2011",
month = "7",
day = "1",
doi = "10.1109/TCSVT.2011.2133570",
language = "English",
volume = "21",
pages = "932--945",
journal = "IEEE Transactions on Circuits and Systems for Video Technology",
issn = "1051-8215",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "7",

}

TY - JOUR

T1 - A network of dynamic probabilistic models for human interaction analysis

AU - Suk, Heung-Il

AU - Jain, Anil K.

AU - Lee, Seong Whan

PY - 2011/7/1

Y1 - 2011/7/1

N2 - We propose a novel method of analyzing human interactions based on the walking trajectories of human subjects, which provide elementary and necessary components for understanding and interpretation of complex human interactions in visual surveillance tasks. Our principal assumption is that an interaction episode is composed of meaningful small unit interactions, which we call sub-interactions. We model each sub-interaction by a dynamic probabilistic model and propose a modified factorial hidden Markov model (HMM) with factored observations. The complete interaction is represented with a network of dynamic probabilistic models (DPMs) by an ordered concatenation of sub-interaction models. The rationale for this approach is that it is more effective in utilizing common components, i.e., sub-interaction models, to describe complex interaction patterns. By assembling these sub-interaction models in a network, possibly with a mixture of different types of DPMs, such as standard HMMs, variants of HMMs, dynamic Bayesian networks, and so on, we can design a robust model for the analysis of human interactions. We show the feasibility and effectiveness of the proposed method by analyzing the structure of network of DPMs and its success on four different databases: a self-collected dataset, Tsinghua University's dataset, the public domain CAVIAR dataset, and the Edinburgh Informatics Forum Pedestrian dataset.

AB - We propose a novel method of analyzing human interactions based on the walking trajectories of human subjects, which provide elementary and necessary components for understanding and interpretation of complex human interactions in visual surveillance tasks. Our principal assumption is that an interaction episode is composed of meaningful small unit interactions, which we call sub-interactions. We model each sub-interaction by a dynamic probabilistic model and propose a modified factorial hidden Markov model (HMM) with factored observations. The complete interaction is represented with a network of dynamic probabilistic models (DPMs) by an ordered concatenation of sub-interaction models. The rationale for this approach is that it is more effective in utilizing common components, i.e., sub-interaction models, to describe complex interaction patterns. By assembling these sub-interaction models in a network, possibly with a mixture of different types of DPMs, such as standard HMMs, variants of HMMs, dynamic Bayesian networks, and so on, we can design a robust model for the analysis of human interactions. We show the feasibility and effectiveness of the proposed method by analyzing the structure of network of DPMs and its success on four different databases: a self-collected dataset, Tsinghua University's dataset, the public domain CAVIAR dataset, and the Edinburgh Informatics Forum Pedestrian dataset.

KW - Dynamic Bayesian network

KW - human interaction analysis

KW - network of dynamic probabilistic models

KW - sub-interactions

KW - video surveillance

UR - http://www.scopus.com/inward/record.url?scp=79960181989&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79960181989&partnerID=8YFLogxK

U2 - 10.1109/TCSVT.2011.2133570

DO - 10.1109/TCSVT.2011.2133570

M3 - Article

AN - SCOPUS:79960181989

VL - 21

SP - 932

EP - 945

JO - IEEE Transactions on Circuits and Systems for Video Technology

JF - IEEE Transactions on Circuits and Systems for Video Technology

SN - 1051-8215

IS - 7

M1 - 5740319

ER -