A Small Sample Face Recognition Method Based on Deep Learning
Meijin Lin, Zihan Zhang, Weijia Zheng · 2020
Face recognition is a hot topic in the field of artificial intelligence in recent years. In many practical applications, face data acquisition is difficult, the accuracy of face recognition based on small samples can't reach a satisfactory accuracy. To solve this problem, a lightweight convolutional neural network (LWCNN) base on deep learning is built for small sample data. In LWCNN, two convolutional blocks are used to extract the features of the input face image. Then the k-fold cross validation is adopted to verify the robustness of the network. The simulation results show that compared with other three methods, our proposed LWCNN achieves better recognition accuracy in small sample space and avoids the over-fitting problem to a certain extent.