AAAI Conference 1999 Conference Paper
ARGUS: An Automated Multi-Agent Visitor Identification System
- Rahul Sukthankar
- Carnegie Mellon University;
- Robert G. Stockton
- Just Research
ARGUS is a multi-agent visitor identification system distributed over several workstations. Human faces are extracted from security cameraimagesby a neuralnetwork-based face detector, and identified as frequent visitors by ARENA, a memory-basedface recognition system. ARGUS then uses a messaging system to notify hosts that their guests have arrived. Aninterface agent enables users to submit feedback, whichis immediately incorporated by ARENA to improve its face recognition performance. The ARGUS components were rapidly developed using JGram, an agent frameworkthat is also detailed in this paper. JGram automatically converts high-level agent specifications into Java source code, and assembles complextasks by composingindividual agent services into a JGram pipeline. ARGUS has been operating successfully in an outdoor environmentfor several months.