RaiseAuth: a novel bio-behavioral authentication method based on ultra-low-complexity movement

Abstract Authentication plays an important role in maintaining social security. Modern authentication methods often relies on mass data datasets to implement by data-driven. However, essential question still remains unclear at level. To what extent can the movement be simplified? We theoretically explain rationality of through arm movements mathematical modeling and design simplest scheme movement. At same time, we collect a small-sample multi-category dataset that compresses as much possible according model function. On this basis, propose method which consists five different cells. Each cell is matched with custom preprocessing module structure. Four cells are composed neural network modules based residual blocks, last traditional machine learning algorithms. The experimental results show also maintain high-accuracy multi-class very simple