A framework for recognizing multi-agent action from visual evidence
AAAI '99/IAAI '99 Proceedings of the sixteenth national conference on Artificial intelligence and the eleventh Innovative applications of artificial intelligence conference innovative applications of artificial intelligence
Multiple view geometry in computer vision
Multiple view geometry in computer vision
GameBots: a flexible test bed for multiagent team research
Communications of the ACM - Internet abuse in the workplace and Game engines in scientific research
Computer Vision
Epipolar Geometry in Stereo, Motion, and Object Recognition: A Unified Approach
Epipolar Geometry in Stereo, Motion, and Object Recognition: A Unified Approach
Recognizing Probabilistic Opponent Movement Models
RoboCup 2001: Robot Soccer World Cup V
Pattern Classification (2nd Edition)
Pattern Classification (2nd Edition)
Robust agent teams via socially-attentive monitoring
Journal of Artificial Intelligence Research
Trajectory based assessment of coordinated human activity
ICVS'03 Proceedings of the 3rd international conference on Computer vision systems
Tracking dynamic team activity
AAAI'96 Proceedings of the thirteenth national conference on Artificial intelligence - Volume 1
Automatic annotation of team actions in observations of embodied agents
Proceedings of the 6th international joint conference on Autonomous agents and multiagent systems
Policy recognition for multi-player tactical scenarios
Proceedings of the 6th international joint conference on Autonomous agents and multiagent systems
Conditional random fields for activity recognition
Proceedings of the 6th international joint conference on Autonomous agents and multiagent systems
Self-organizing social and spatial networks under what-if scenarios
Proceedings of the 6th international joint conference on Autonomous agents and multiagent systems
Monitoring human behavior in an assistive environment using multiple views
Proceedings of the 1st international conference on PErvasive Technologies Related to Assistive Environments
A novel sequence representation for unsupervised analysis of human activities
Artificial Intelligence
Multi-agent plan adaptation using coordination patterns in team adversarial games
Proceedings of the 9th International Conference on Autonomous Agents and Multiagent Systems: volume 1 - Volume 1
Recognising Agent Behaviour During Variable Length Activities
Proceedings of the 2010 conference on ECAI 2010: 19th European Conference on Artificial Intelligence
Conditional random fields for activity recognition in smart environments
Proceedings of the 1st ACM International Health Informatics Symposium
Market-based dynamic task allocation using heuristically accelerated reinforcement learning
EPIA'11 Proceedings of the 15th Portugese conference on Progress in artificial intelligence
Agent-oriented incremental team and activity recognition
IJCAI'11 Proceedings of the Twenty-Second international joint conference on Artificial Intelligence - Volume Volume Two
Detecting and identifying coalitions
Proceedings of the 11th International Conference on Autonomous Agents and Multiagent Systems - Volume 3
Learning collaborative team behavior from observation
Expert Systems with Applications: An International Journal
Hi-index | 0.00 |
This paper presents a framework for robustly recognizing physical team behaviors by exploiting spatio-temporal patterns. Agent team behaviors in athletic and military domains typically exhibit an observable structure characterized by the relative positions of teammates and external landmarks, such as a team of soldiers ambushing an opponent or a soccer player moving to receive a pass. We demonstrate how complex team relationships that are not easily expressed by region-based heuristics can be modeled from data and domain knowledge in a way that is robust to noise and spatial variation. To represent team behaviors in our domain of MOUT (Military Operations in Urban Terrain) planning, we employ two classes of spatial models: 1) team templates that encode static relationships between team members and external landmarks; and 2) spatially-invariant Hidden Markov Models (HMMs) to represent evolving agent team configurations over time. These two classes of models can be combined to improve recognition accuracy, particularly for behaviors that appear similar in static snapshots. We evaluate our modeling techniques on large urban maps and position traces of two-person human teams performing MOUT behaviors in a customized version of Unreal Tournament (a commercially available first-person shooter game).