A Bayesian computer vision system for modeling human interactions

TLDR

Interaction behaviors such as following, path alteration, and meeting are key targets for surveillance. The study aims to develop and compare real‑time computer‑vision systems that model and recognize human behaviors using HMM and CHMM architectures. The system integrates top‑down and bottom‑up Bayesian components in a closed feedback loop, employs HMM/CHMM state‑based learning, and uses a synthetic Alife‑style training system to generate prior models. The Bayesian models accurately classify real human behaviors and interactions without further tuning or training.

Abstract

We describe a real-time computer vision and machine learning system for modeling and recognizing human behaviors in a visual surveillance task. The system deals in particularly with detecting when interactions between people occur and classifying the type of interaction. Examples of interesting interaction behaviors include following another person, altering one's path to meet another, and so forth. Our system combines top-down with bottom-up information in a closed feedback loop, with both components employing a statistical Bayesian approach. We propose and compare two different state-based learning architectures, namely, HMMs and CHMMs for modeling behaviors and interactions. Finally, a synthetic "Alife-style" training system is used to develop flexible prior models for recognizing human interactions. We demonstrate the ability to use these a priori models to accurately classify real human behaviors and interactions with no additional tuning or training.