Physiological and behavioural approaches for person recognition using videos

Abstract

In this article we present two physiological and behavioural approaches for person recognition using videos. The first system, called the multimodal recognition system, is divided in two modules. The first module exploits the behavioural information: it is based on statistical features computed using the displacement signals of the head; the second one is dealing with the physiological information: it is a probabilistic extension of the classic Eigenface approach. For a consistent fusion, both systems share the same probabilistic classification framework: a Gaussian Mixture Model (GMM) approximation and a Bayesian classifier. The second system, called the tomoface recognition system, applies discrete video tomography to compute spatiotemporal features that summarise the head and facial dynamics of a sequence into a single image (called "video X-ray image"); these novel features are subsequently analysed by an extended version of the eigenface approach. Finally, we assess the performances of both systems, and we compare them with a traditional recognition approach based on facial appearance.