Abstract:

To reduce the memory occupancy rate and computational resources in face recognition with high-dimension features extracted from large convolutional neural networks, an efficient Fully Convolutional Network (FCN) of the deep hash combined with global and local pooling. First, an FCN based on Global Average Pooling (GAP) is proposed to reduce network parameters and compress the model size. Second, a fusion method for learning approximate hash coding with multiple classification properties is used with Quantization Loss and Softmax Loss. Experimental results show that the method proposed can improve the efficiency up to 68% and that the Rank-1 accuracy is increased slightly with the Visual Geometry Group (VGG) framework. The fusion loss method can improve the efficiency up to 23.7% and the Rank-1 accuracy is maintained with the Face Residual Network (Face-ResNet) framework. The results indicate that the proposed method can improve the efficiency both from feature extraction and reduction. It also can be applied to other frameworks.