Inferring Leadership from Group Dynamics Using Markov Chain Monte Carlo Methods

Avishy Carmi 1 Lyudmila Mihaylova 2 François Septier 3, 4 Sze Kim Pang 5 Pini Gurfil 6 Simon J. Godsill 7
1 Department of Mechanical Engineering [Beer-Sheva]
2 Department of Automatic Control and Systems Engineering
3 LAGIS-SI
LAGIS - Laboratoire d'Automatique, Génie Informatique et Signal
4 Institut TELECOM/TELECOM Lille1
5 DSO National Laboratories
6 Faculty of Aerospace Engineering
7 Signal Processing Group Dept of Engineering
Abstract : This chapter presents a novel framework for identifying and tracking dominant agents in groups. The proposed approach relies on a causality detection scheme that is capable of ranking agents with respect to their contribution in recognizing the system's collective behavior based exclusively on the agents' observed trajectories. Further, the reasoning paradigm is made robust to multiple emissions and clutter by employing a class of recently introduced Markov chain Monte Carlo-based group tracking methods. Examples are provided that demonstrate the strong potential of the proposed scheme in identifying actual leaders in swarms of interacting agents and moving crowds.
Document type :
Book sections
Complete list of metadatas
https://hal-imt.archives-ouvertes.fr/hal-00937296
Contributor : François Septier <>
Submitted on : Tuesday, January 28, 2014 - 10:52:53 AM
Last modification on : Thursday, June 13, 2019 - 11:14:02 AM

Identifiers

  • HAL Id : hal-00937296, version 1

Collections

INSTITUT-TELECOM | LAGIS | LAGIS-SI | IMT-LILLE-DOUAI

Citation

Avishy Carmi, Lyudmila Mihaylova, François Septier, Sze Kim Pang, Pini Gurfil, et al.. Inferring Leadership from Group Dynamics Using Markov Chain Monte Carlo Methods. Modeling, Simulation and Visual Analysis of Crowds: A Multidisciplinary Perspective, Eds S. Ali, K. Nishino, D. Manocha, M. Shah - Springer, pp.325-346, 2013, The International Series in Video Computing Volume 11, 978-1461484820. ⟨hal-00937296⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

202

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD