AFace: Range-flexible Anti-spoofing Face Authentication via Smartphone Acoustic Sensing

User authentication on smartphones needs to balance both security and convenience. Many image-based face authentication methods are vulnerable to spoofing and are plagued by privacy breaches, so models based on acoustic sensing have emerged to achieve reliable user authentication. However, they can only achieve reasonable performance under specific conditions (i.e., a fixed range), and they can not resist 3D printing attacks. To address these limitations, we present a novel user authentication system, referred to as AFace. The system mainly consists of two parts: an iso-depth model and a range-adaptive (RA) algorithm. The iso-depth model establishes a connection between acoustic echoes and facial structures, while taking into account the influence of biological materials on echo energy, making it resistant to 3D printing attacks (as it's difficult to replicate material information in 3D printing). RA algorithm can adaptively compensate for the distance between the user and the smartphone, enabling flexible authentication modes. Results from experiments with 40 volunteers demonstrate that AFace achieves an average accuracy of 96.9% and an F1 score of 96.9%, and no image/video-based attack is observed to succeed in spoofing.