TitleABROA : Audio-based Room-occupancy Analysis using Gaussian Mixtures and Hidden Markov Models
Publication TypeConference Paper
Year of Publication2016
AuthorsValle, R
Conference Name Future Technologies Conference (FTC) 2016
Conference LocationSan Francisco
Abstract

This paper outlines preliminary steps towards the development of an audio used room occupancy analysis model. Our approach borrows fromthe speech recognition tradition and is based on GaussianMixture and a Hidden Markov Model. 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 fromexperiments with audio data from a retail store in Palo Alto, California. Model assessment is done via leave-one-out Bootstrap anand model convergence achieves good accuracy, thus representing a contribution to cross-modal people counting algorithms.