Face recognition method and device, electronic equipment and storage medium

Abstract

The invention relates to the technical field of human face recognition, in particular to a human face recognition method and device, electronic equipment and a storage medium. The human face recognition method adopts a deep learning network model to recognize human face information in a sample image, and the human face recognition method is characterized by comprising the following steps: the human face information in the sample image is recognized; constructing corresponding probability distribution tensors for a plurality of channel groups respectively, the plurality of channel groups beingobtained by segmenting channel dimensions according to a preset group number, and the channel groups having different weights; selecting a convolution kernel corresponding to the maximum channel group, and adaptively adjusting feature information extracted from different sample images; wherein the maximum channel group is the channel group with the highest response in each constructed probabilitydistribution tensor; and carrying out convolution on each channel group according to the convolution kernel to obtain convolution features, and the obtained multiple convolution features are spliced.According to the embodiment of the invention, the feature expression capability of the network model is improved, and the complexity of the whole network structure is not increased.

Description

technical field [0001] The present invention relates to the technical field of face recognition, in particular to a face recognition method, device, electronic equipment and storage medium. Background technique [0002] Face recognition technology is one of the important research directions in the field of computer vision, which mainly achieves the purpose of identifying identity by analyzing and comparing facial features. Because face features can be collected in a non-contact manner, it has the advantages of simplicity and convenience. These advantages make face recognition technology stand out among many biometric technologies. The market share of landing products is large, and it is widely used in security, economy, etc. in the field. [0003] In 2012, the AlexNet network model was proposed, demonstrating the amazing accuracy advantages of deep neural networks in the direction of image classification, allowing researchers at home and abroad to see new directions, contin...

AI Technical Summary

Problems solved by technology

However, the convolution operation is parameter-shared in the network structure. If you want to obtain richer feature information on different samples, you need to increase the number of convolutional layers, which not only increases the amount of parameter calculation, but also makes it difficult to optimize the network model.

Images