Multi-Agent Event Recognition

Somboon Hongeng


In this talk, we will present a new approach to recognizing multi-agent events observed by a static camera. We address both the problems of tracking multiple objects and of recognizing the events from uncertain data. To track objects robustly, knowledge about the ground plane and the events is used.

An event is considered as composed of action threads, each thread being executed by a single actor. A single thread of action is recognized from the characteristics of the trajectory and moving blob of the actor using Bayesian methods. A multi-agent event is represented by a number of action threads related by temporal constraints. Multi-agent events are recognized by propagating the constraints and likelihoods of event threads in a temporal logic network. We present results on real-world data.

Maintained by Philippos Mordohai