Footstep recognition for a smart home environment

This paper reports some experiments which assess the potential use of a footstep biometric verification system for a smart home environment. We present a semi-automatic capture system and report results on a database with independent development and evaluation datasets comprised of more than 3500 footsteps collected from 55 persons. We present an optimisation of geometric and holistic feature extraction approaches. An equal error rate of 13% is obtained with holistic features classified with a support vector machine. The database is freely available to the research community.