ABROA : Audio-Based Room-Occupancy Analysis using Gaussian Mixtures and Hidden Markov Models

TL;DR

This paper presents an audio-based room occupancy analysis method using Gaussian Mixtures and Hidden Markov Models, demonstrating promising results in a retail environment and contributing to multimodal people counting techniques.

Contribution

It introduces a novel audio-based occupancy analysis model leveraging speech recognition techniques, with solutions for feature design and prediction, validated through experiments.

Findings

Good model convergence and accuracy

Effective feature design strategies

Potential for multimodal people counting

Abstract

This paper outlines preliminary steps towards the development of an audio- based room-occupancy analysis model. Our approach borrows from speech recognition tradition and is based on Gaussian Mixtures and Hidden Markov Models. We analyze possible challenges encountered in the development of such a model, and offer several solutions including feature design and prediction strategies. We provide results obtained from experiments with audio data from a retail store in Palo Alto, California. Model assessment is done via leave-two-out Bootstrap and model convergence achieves good accuracy, thus representing a contribution to multimodal people counting algorithms.