Learning models of activities involving interacting objects

Cristina Manfredotti; Kim Steenstrup Pedersen; Howard J. Hamilton; Sandra Zilles

doi:10.1007/978-3-642-41398-8_25

Learning models of activities involving interacting objects

Cristina Manfredotti, Kim Steenstrup Pedersen, Howard J. Hamilton, Sandra Zilles

Department of Computer Science

1 Citation (Scopus)

53 Downloads (Pure)

Abstract

We propose the LEMAIO multi-layer framework, which makes use of hierarchical
abstraction to learn models for activities involving multiple interacting objects
from time sequences of data concerning the individual objects.
Experiments in the sea navigation domain yielded learned models that were then successfully applied to activity recognition, activity simulation and multi-target tracking. Our method compares favourably with respect to previously reported results using Hidden Markov Models and Relational Particle Filtering.

Original language	English
Title of host publication	Advances in Intelligent Data Analysis XII : 12th International Symposium, IDA 2013, London, UK, October 17-19, 2013, Proceedings
Editors	Allan Tucker, Frank Höppner, Arno Siebes, Stephen Swift
Number of pages	13
Publisher	Springer
Publication date	2013
Pages	285-297
ISBN (Print)	978-3-642-41397-1
ISBN (Electronic)	978-3-642-41398-8
DOIs	https://doi.org/10.1007/978-3-642-41398-8_25
Publication status	Published - 2013
Event	12th International Symposium on Advances in Intelligent Data Analysis - London, United Kingdom Duration: 17 Oct 2013 → 19 Oct 2013 Conference number: 12

Conference

Conference	12th International Symposium on Advances in Intelligent Data Analysis
Number	12
Country/Territory	United Kingdom
City	London
Period	17/10/2013 → 19/10/2013

Series	Lecture notes in computer science
Volume	8207

Access to Document

10.1007/978-3-642-41398-8_25

Learning models of activities involving interacting objectsAccepted author manuscript, 804 KBLicence: Other

Cite this

Manfredotti, C., Pedersen, K. S., Hamilton, H. J., & Zilles, S. (2013). Learning models of activities involving interacting objects. In A. Tucker, F. Höppner, A. Siebes, & S. Swift (Eds.), Advances in Intelligent Data Analysis XII: 12th International Symposium, IDA 2013, London, UK, October 17-19, 2013, Proceedings (pp. 285-297). Springer. https://doi.org/10.1007/978-3-642-41398-8_25

Learning models of activities involving interacting objects. / Manfredotti, Cristina; Pedersen, Kim Steenstrup; Hamilton, Howard J. et al.
Advances in Intelligent Data Analysis XII: 12th International Symposium, IDA 2013, London, UK, October 17-19, 2013, Proceedings. ed. / Allan Tucker; Frank Höppner; Arno Siebes; Stephen Swift. Springer, 2013. p. 285-297 (Lecture notes in computer science, Vol. 8207).

Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review

Manfredotti, C, Pedersen, KS, Hamilton, HJ & Zilles, S 2013, Learning models of activities involving interacting objects. in A Tucker, F Höppner, A Siebes & S Swift (eds), Advances in Intelligent Data Analysis XII: 12th International Symposium, IDA 2013, London, UK, October 17-19, 2013, Proceedings. Springer, Lecture notes in computer science, vol. 8207, pp. 285-297, 12th International Symposium on Advances in Intelligent Data Analysis, London, United Kingdom, 17/10/2013. https://doi.org/10.1007/978-3-642-41398-8_25

Manfredotti C, Pedersen KS, Hamilton HJ, Zilles S. Learning models of activities involving interacting objects. In Tucker A, Höppner F, Siebes A, Swift S, editors, Advances in Intelligent Data Analysis XII: 12th International Symposium, IDA 2013, London, UK, October 17-19, 2013, Proceedings. Springer. 2013. p. 285-297. (Lecture notes in computer science, Vol. 8207). doi: 10.1007/978-3-642-41398-8_25

Manfredotti, Cristina ; Pedersen, Kim Steenstrup ; Hamilton, Howard J. et al. / Learning models of activities involving interacting objects. Advances in Intelligent Data Analysis XII: 12th International Symposium, IDA 2013, London, UK, October 17-19, 2013, Proceedings. editor / Allan Tucker ; Frank Höppner ; Arno Siebes ; Stephen Swift. Springer, 2013. pp. 285-297 (Lecture notes in computer science, Vol. 8207).

@inproceedings{237d389379ae4fcab3c41bf0f3d7ed49,

title = "Learning models of activities involving interacting objects",

abstract = "We propose the LEMAIO multi-layer framework, which makes use of hierarchicalabstraction to learn models for activities involving multiple interacting objectsfrom time sequences of data concerning the individual objects. Experiments in the sea navigation domain yielded learned models that were then successfully applied to activity recognition, activity simulation and multi-target tracking. Our method compares favourably with respect to previously reported results using Hidden Markov Models and Relational Particle Filtering.",

author = "Cristina Manfredotti and Pedersen, {Kim Steenstrup} and Hamilton, {Howard J.} and Sandra Zilles",

year = "2013",

doi = "10.1007/978-3-642-41398-8_25",

language = "English",

isbn = "978-3-642-41397-1",

series = "Lecture notes in computer science",

publisher = "Springer",

pages = "285--297",

editor = "Allan Tucker and Frank H{\"o}ppner and Arno Siebes and Stephen Swift",

booktitle = "Advances in Intelligent Data Analysis XII",

note = "12th International Symposium on Advances in Intelligent Data Analysis, IDA 2013 ; Conference date: 17-10-2013 Through 19-10-2013",

}

TY - GEN

T1 - Learning models of activities involving interacting objects

AU - Manfredotti, Cristina

AU - Pedersen, Kim Steenstrup

AU - Hamilton, Howard J.

AU - Zilles, Sandra

N1 - Conference code: 12

PY - 2013

Y1 - 2013

N2 - We propose the LEMAIO multi-layer framework, which makes use of hierarchicalabstraction to learn models for activities involving multiple interacting objectsfrom time sequences of data concerning the individual objects. Experiments in the sea navigation domain yielded learned models that were then successfully applied to activity recognition, activity simulation and multi-target tracking. Our method compares favourably with respect to previously reported results using Hidden Markov Models and Relational Particle Filtering.

AB - We propose the LEMAIO multi-layer framework, which makes use of hierarchicalabstraction to learn models for activities involving multiple interacting objectsfrom time sequences of data concerning the individual objects. Experiments in the sea navigation domain yielded learned models that were then successfully applied to activity recognition, activity simulation and multi-target tracking. Our method compares favourably with respect to previously reported results using Hidden Markov Models and Relational Particle Filtering.

U2 - 10.1007/978-3-642-41398-8_25

DO - 10.1007/978-3-642-41398-8_25

M3 - Article in proceedings

SN - 978-3-642-41397-1

T3 - Lecture notes in computer science

SP - 285

EP - 297

BT - Advances in Intelligent Data Analysis XII

A2 - Tucker, Allan

A2 - Höppner, Frank

A2 - Siebes, Arno

A2 - Swift, Stephen

PB - Springer

T2 - 12th International Symposium on Advances in Intelligent Data Analysis

Y2 - 17 October 2013 through 19 October 2013

ER -