Bayesian inference of recursive sequences of group activities from tracks

Ernesto Brau, Colin Dawson, Alfredo Carrillo, David Sidi, Clayton T Morrison

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We present a probabilistic generative model for inferring a description of coordinated, recursively structured group activities at multiple levels of temporal granularity based on observations of individuals' trajectories. The model accommodates: (1) hierarchically structured groups, (2) activities that are temporally and compositionally recursive, (3) component roles assigning different subactivity dynamics to subgroups of participants, and (4) a nonparametric Gaussian Process model of trajectories. We present an MCMC sampling framework for performing joint inference over recursive activity descriptions and assignment of trajectories to groups, integrating out continuous parameters. We demonstrate the model's expressive power in several simulated and complex real-world scenarios from the VIRAT and UCLA Aerial Event video data sets.

Original languageEnglish (US)
Title of host publication30th AAAI Conference on Artificial Intelligence, AAAI 2016
PublisherAAAI press
Pages1129-1137
Number of pages9
ISBN (Electronic)9781577357605
StatePublished - 2016
Event30th AAAI Conference on Artificial Intelligence, AAAI 2016 - Phoenix, United States
Duration: Feb 12 2016Feb 17 2016

Other

Other30th AAAI Conference on Artificial Intelligence, AAAI 2016
CountryUnited States
CityPhoenix
Period2/12/162/17/16

Fingerprint

Trajectories
Antennas
Sampling
Statistical Models

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Brau, E., Dawson, C., Carrillo, A., Sidi, D., & Morrison, C. T. (2016). Bayesian inference of recursive sequences of group activities from tracks. In 30th AAAI Conference on Artificial Intelligence, AAAI 2016 (pp. 1129-1137). AAAI press.

Bayesian inference of recursive sequences of group activities from tracks. / Brau, Ernesto; Dawson, Colin; Carrillo, Alfredo; Sidi, David; Morrison, Clayton T.

30th AAAI Conference on Artificial Intelligence, AAAI 2016. AAAI press, 2016. p. 1129-1137.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Brau, E, Dawson, C, Carrillo, A, Sidi, D & Morrison, CT 2016, Bayesian inference of recursive sequences of group activities from tracks. in 30th AAAI Conference on Artificial Intelligence, AAAI 2016. AAAI press, pp. 1129-1137, 30th AAAI Conference on Artificial Intelligence, AAAI 2016, Phoenix, United States, 2/12/16.
Brau E, Dawson C, Carrillo A, Sidi D, Morrison CT. Bayesian inference of recursive sequences of group activities from tracks. In 30th AAAI Conference on Artificial Intelligence, AAAI 2016. AAAI press. 2016. p. 1129-1137
Brau, Ernesto ; Dawson, Colin ; Carrillo, Alfredo ; Sidi, David ; Morrison, Clayton T. / Bayesian inference of recursive sequences of group activities from tracks. 30th AAAI Conference on Artificial Intelligence, AAAI 2016. AAAI press, 2016. pp. 1129-1137
@inproceedings{d5cd83f89cfc4c31ad304a18d8b4b8b8,
title = "Bayesian inference of recursive sequences of group activities from tracks",
abstract = "We present a probabilistic generative model for inferring a description of coordinated, recursively structured group activities at multiple levels of temporal granularity based on observations of individuals' trajectories. The model accommodates: (1) hierarchically structured groups, (2) activities that are temporally and compositionally recursive, (3) component roles assigning different subactivity dynamics to subgroups of participants, and (4) a nonparametric Gaussian Process model of trajectories. We present an MCMC sampling framework for performing joint inference over recursive activity descriptions and assignment of trajectories to groups, integrating out continuous parameters. We demonstrate the model's expressive power in several simulated and complex real-world scenarios from the VIRAT and UCLA Aerial Event video data sets.",
author = "Ernesto Brau and Colin Dawson and Alfredo Carrillo and David Sidi and Morrison, {Clayton T}",
year = "2016",
language = "English (US)",
pages = "1129--1137",
booktitle = "30th AAAI Conference on Artificial Intelligence, AAAI 2016",
publisher = "AAAI press",

}

TY - GEN

T1 - Bayesian inference of recursive sequences of group activities from tracks

AU - Brau, Ernesto

AU - Dawson, Colin

AU - Carrillo, Alfredo

AU - Sidi, David

AU - Morrison, Clayton T

PY - 2016

Y1 - 2016

N2 - We present a probabilistic generative model for inferring a description of coordinated, recursively structured group activities at multiple levels of temporal granularity based on observations of individuals' trajectories. The model accommodates: (1) hierarchically structured groups, (2) activities that are temporally and compositionally recursive, (3) component roles assigning different subactivity dynamics to subgroups of participants, and (4) a nonparametric Gaussian Process model of trajectories. We present an MCMC sampling framework for performing joint inference over recursive activity descriptions and assignment of trajectories to groups, integrating out continuous parameters. We demonstrate the model's expressive power in several simulated and complex real-world scenarios from the VIRAT and UCLA Aerial Event video data sets.

AB - We present a probabilistic generative model for inferring a description of coordinated, recursively structured group activities at multiple levels of temporal granularity based on observations of individuals' trajectories. The model accommodates: (1) hierarchically structured groups, (2) activities that are temporally and compositionally recursive, (3) component roles assigning different subactivity dynamics to subgroups of participants, and (4) a nonparametric Gaussian Process model of trajectories. We present an MCMC sampling framework for performing joint inference over recursive activity descriptions and assignment of trajectories to groups, integrating out continuous parameters. We demonstrate the model's expressive power in several simulated and complex real-world scenarios from the VIRAT and UCLA Aerial Event video data sets.

UR - http://www.scopus.com/inward/record.url?scp=85007271672&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85007271672&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85007271672

SP - 1129

EP - 1137

BT - 30th AAAI Conference on Artificial Intelligence, AAAI 2016

PB - AAAI press

ER -