Age Estimation and Gender Classification Based on Human Gait Analysis

Human gait is an important biometric trait which is used in a variety of applications. The development of low-cost inertial sensors has facilitated the capturing of human gait, widening its use in commercial applications. This study proposes a method for gender classification and age estimation using the largest available inertial sensors human gait dataset. The proposed method has three stages. Identification of human gait cycles from inertial sensors signals is performed at the first stage. At the second stage, statistical features are extracted from identified gait cycles. The features extracted at the second stage are used in various machine learning algorithms for gender classification and age estimation at the third stage. A series of experiments are performed, and results show the effectiveness and promising potential of the proposed method for age estimation and gender classification.