Research on the Face Silent Liveness Algorithm Based on Auxiliary Information

With the rapid development of artificial intelligence and computer vision, face recognition system has been widely used in the field of internet finance to achieve digital authentication and the security of systems has received attention increasingly. As an important part of the face recognition system which is used to resist the attack, the liveness detection will directly affect the security level of the whole system. In order to improve the existing silent liveness algorithms with shortcomings, such as insufficient features, over-fitting easily and lacking of interpretability, this paper proposes a face silent liveness detection algorithm based on multi -dimension auxiliary information and the multi -dimension information, such as depth map, reflection map and spectral map, are introduced as the auxiliary supervision in the liveness algorithm. In order to verify the validity of the proposed liveness detection algorithm, multiple comparative experiments are set up and different kinds of auxiliary information are introduced for comparisons. All experiments are conducted on the CelebA-Spoof data set. The experimental results are highly consistent with the theoretical expectations of the algorithm design, which provides more interpretability for the liveness detection.