L.J. Luotsinen and L. Bölöni

Role-Based Teamwork Activity Recognition in Observations of Embodied Agent Actions


Cite as:

L.J. Luotsinen and L. Bölöni. Role-Based Teamwork Activity Recognition in Observations of Embodied Agent Actions. In The Seventh Intl. Joint Conf. on Autonomous Agents and Multi-Agent Systems (AAMAS 08), pp. 567–574, 2008.

Download:

Download 

Abstract:

Recognizing team actions in the behavior of embodied agents has many practical applications and had seen significant progress in recent years. One approach with proven results is based on HMM-based recognition of spatio-temporal patterns in the behavior of the agents. While it had been shown to work on real-world datasets, this approach was found to be brittle. In this paper we present two contributions which together can significantly increase the robustness of teamwork activity recognition. First we introduce a technique to reduce high dimensional continuous input data to a set of discrete features, which capture the essential components of the team actions. Second, we prefix the actual team action recognition with a role recognition module, which allows us to present the recognizer with arbitrarily shuffled input, and still obtain high recognition rates. We validate the improved accuracy and robustness of the team action recognizer on datasets derived from captured real world data.

BibTeX:

@inproceedings{Luotsinen-2008-AAMAS,
author = "L.J. Luotsinen and L. B{\"o}l{\"o}ni",
title = "Role-Based Teamwork Activity Recognition in Observations of Embodied Agent Actions",
booktitle = "The Seventh Intl. Joint Conf. on Autonomous Agents and Multi-Agent Systems (AAMAS 08)",
year = "2008",
pages = "567-574",
abstract = {
  Recognizing team actions in the behavior of embodied agents has many
  practical applications and had seen significant progress in recent years.
  One approach with proven results is based on HMM-based recognition of
  spatio-temporal patterns in the behavior of the agents. While it had been
  shown to work on real-world datasets, this approach was found to be
  brittle.
  In this paper we present two contributions which together can
  significantly increase the robustness of teamwork activity recognition.
  First we introduce a technique to reduce high dimensional continuous
  input data to a set of discrete features, which capture the essential
  components of the team actions. Second, we prefix the actual team action
  recognition with a role recognition module, which allows us to present
  the recognizer with arbitrarily shuffled input, and still obtain high
  recognition rates.
  We validate the improved accuracy and robustness of the team action
  recognizer on datasets derived from captured real world data.
 },
xnote = "acceptance rate 142 / 640 = 22\%",
}

Generated by bib2html.pl (written by Patrick Riley, Lotzi Boloni ) on Fri Jan 29, 2021 20:15:21