Individuals, Groups, and Crowds: Modelling Complex, Multi-Object Behaviour in Phase Space. Sethi, J, R., Roy-Chowdhury, & K, A. In ICCV VECTaR, 2011.
abstract   bibtex   
This paper concentrates on the problem of modelling and recognition of complex behaviours involving multi-object interactions in video. We use motion patterns of individual objects to construct models which characterize pairs by cor- relating them in phase space. These models of complex in- teractions allow for: recognition of group activities, which occur when individual people or objects start interacting as a single entity; detection of transitions from individu- als to groups to crowds; and the interactions of individu- als with groups, as well as the interactions of groups with other groups. We establish a general formalism by exam- ining activities using relative distances and analyse multi- object interactions directly via the Phase Space Algorithm. Finally, we calculate a scale-invariant Group Transition Ratio to quantify formation and dispersal of both groups and crowds. Our input is solely the position information of individuals, which we get using a person tracker, optical flow, and Lagrangian particle dynamics. We demonstrate the uses of this model for recognition of complex activi- ties on the standard CAVIAR, VIVID, and UCR Videoweb datasets.
@inproceedings{ Sethi2011a,
  abstract = {This paper concentrates on the problem of modelling and recognition of complex behaviours involving multi-object interactions in video. We use motion patterns of individual objects to construct models which characterize pairs by cor- relating them in phase space. These models of complex in- teractions allow for: recognition of group activities, which occur when individual people or objects start interacting as a single entity; detection of transitions from individu- als to groups to crowds; and the interactions of individu- als with groups, as well as the interactions of groups with other groups. We establish a general formalism by exam- ining activities using relative distances and analyse multi- object interactions directly via the Phase Space Algorithm. Finally, we calculate a scale-invariant Group Transition Ratio to quantify formation and dispersal of both groups and crowds. Our input is solely the position information of individuals, which we get using a person tracker, optical flow, and Lagrangian particle dynamics. We demonstrate the uses of this model for recognition of complex activi- ties on the standard CAVIAR, VIVID, and UCR Videoweb datasets.},
  author = {Sethi, Ricky J and Roy-Chowdhury, Amit K},
  booktitle = {ICCV VECTaR},
  file = {:C$\backslash$:/Users/rjs/Documents/Mendeley Desktop/Sethi, Roy-Chowdhury/ICCV VECTaR/Sethi, Roy-Chowdhury_2011_Individuals, Groups, and Crowds Modelling Complex, Multi-Object Behaviour in Phase Space.pdf:pdf},
  title = {{Individuals, Groups, and Crowds: Modelling Complex, Multi-Object Behaviour in Phase Space}},
  year = {2011}
}

Downloads: 0