Adaptive Gait Biometrics for Real Time Continuous Authentication

Abstract

This paper presents a continuous authentication system that leverages gait analysis by integrating real-time image processing and machine learning to overcome the vulnerabilities of traditional one-time verification methods. Utilizing gait - a natural, spoof-resistant biometric - the framework enables non-intrusive, real-time identity validation. A Microsoft Kinect sensor captures 3D kinematic features (speed, stride length, joint angles), which are processed by machine learning models to authenticate users against pre-trained profiles. The system dynamically adapts to environmental variables (lighting, terrain) and mitigates gait variability (e.g., injuries) via confidence-based early stopping, thereby optimizing computational efficiency. Privacy is maintained through encrypted data storage and processing. Evaluations reveal robust performance in controlled settings and scalability for high-security applications such as access control. Although accuracy is promising, challenges persist in heterogeneous environments, underscoring the need for multi-sensor integration. This work advances gait biometrics by balancing security and usability, offering a passive authentication solution that enhances convenience while protecting against impersonation threats.