Gender and age estimation from gait: A review

Abstract

Gait—the way a person walks, is a behavioral biometric which can be used for human identification purpose just like other biometrics—fingerprints, hand geometry, eyes, face, ear, etc. Apart from that, gait can also be used for the estimation of an individual’s gender, age group, and age. Age estimation has relatively more extensive applications like in visual surveillance and monitoring, access control to places like shopping malls, airports, liquor shops, public clubs, etc. Over the years many age estimation methods and techniques have been proposed majorly in the computer vision field. This research depicts a broad literature survey of gender and age estimation techniques from gait. Methodical analysis of age estimation algorithms, their classification, different quantified features and representations of gait used, and their merits and demerits are described. We have also presented a separate analysis of various human gait databases that have been developed over the years in this research area. This study presents a holistic comparative study thereby providing a—way to go ahead, in this research area.