Face recognition neural network adjustment method and device

Abstract

The invention provides a method and a device for adjusting and deploying a face recognition neural network. The face recognition neural network at least comprises a plurality of convolution layers andat least one full connection layer, the last full connection layer is a classifier for classification, and the method comprises the following steps: obtaining a neural network model to be trained; training the neural network model by using fixed-point quantization to obtain a trained fixed-point quantization neural network model in which the last full connection layer maintains a floating point in a training process; and outputting the trained fixed-point quantized neural network model without the last full connection layer. Therefore, by utilizing the particularity of the face recognition network, the classifier layer unfixed points which greatly influence the overall precision of the network are kept in the training stage and are not input into the network any more, and the classifier layer is not included in the network any more, so that the trained fixed-point neural network can be ensured to have high precision, and extra computing power requirements during network deployment canbe avoided at the same time.

Description

Technical field [0001] The present invention relates to deep learning, and in particular to fixed-point quantification of face recognition neural networks. Background technique [0002] When machine learning was born, face recognition was one of the most basic application areas. In recent years, with the continuous development and progress of deep learning, the accuracy of face recognition has also risen rapidly, even exceeding the human level in multiple evaluation sets. Therefore, face recognition has also begun to be applied in life scenes such as smart phones and smart surveillance cameras. [0003] However, the limitations of computing power and power consumption in actual application scenarios severely restrict the promotion and deployment of face recognition technology. In view of the fact that the existing neural network parameters have a large amount of redundancy, the use of resources can be greatly reduced through the fixed point of the neural network. Therefore, how ...

AI Technical Summary

Problems solved by technology

[0003] However, the limitations of computing power and power consumption in actual application scenarios seriously restrict the promotion and deployment of face recognition technology.

Images