Twin neural network training method for face verification

A neural network training and artificial neural network technology, applied in the field of twin neural network training, can solve problems such as difficulty in labeling, reduce workload, improve the effect of feature extraction, and simplify the processing process.

Inactive Publication Date: 2019-01-01
HANGZHOU DIANZI UNIV
View PDF4 Cites 37 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the problem of large-scale face data collection and labeling difficulty in the above-mentioned prior art, and to provide a twin neural network training method for face verification, which can be obtained by using less sample data. Recognition model with high recognition accuracy

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Twin neural network training method for face verification
  • Twin neural network training method for face verification
  • Twin neural network training method for face verification

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] The technical solutions of the present invention will be further specifically described below through specific embodiments and in conjunction with the accompanying drawings.

[0026] A twin neural network training method for face verification, comprising: preparing a training sample set; normalizing the size of the pictures in the training sample set and then inputting them into the artificial neural network for training; the artificial neural network includes two identical Sub-neural network; divide the processed training samples into data sets data and data_p with equal numbers, and send the data sets data and data_p to two sub-neural networks respectively for sample feature vector extraction; realize the neural network by comparing the loss function Iterative optimization until the number of iterations reaches the set value, then jump out of the iteration, at this time the trained artificial neural network is the twin neural network for face verification; the contrast...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

A twin neural network training method for face verification comprises: a training sample set is prepared; the images in the training sample set are normalized in size and then input to the artificialneural network for training. Artificial neural network consists of two identical sub-neural networks. The training samples are divided into datasets data_p and data_p which are equal in number. The datasets data_p and data_p are sent to two sub-neural networks to extract the eigenvectors of the samples. By comparing the loss function to realize the iterative optimization of the neural network, until the iterative number reaches the set value, then jump out of the iteration, at this time the trained artificial neural network is the twin neural network for face verification; The contrast loss function represents similarity between two sets of eigenvectors.

Description

technical field [0001] The invention relates to the technical field related to computer vision, in particular to a twin neural network training method for face verification. Background technique [0002] Face recognition technology, as a kind of recognition based on physiological characteristics in the field of biometric recognition, is a technology that extracts facial features through computers and performs identity verification based on these features. Compared with other biometric technologies, face recognition is a non-contact recognition technology, which has the advantages of fast, simple, accurate and reliable, cost-effective, and good scalability. At present, it is widely used in file management, security, mobile payment and other fields. [0003] However, the accuracy of face recognition is often affected by four factors: illumination, posture, expression and age changes, making image-based face recognition still face a series of challenges. Among them, age chang...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V40/168G06N3/045G06F18/2413
Inventor 常昊孔亚广刘威屠雨泽
Owner HANGZHOU DIANZI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap