Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Face recognition method based on deep neural network

A deep neural network and face recognition technology, applied in biological neural network model, neural architecture, character and pattern recognition, etc., can solve the problem of consuming GPU memory, achieve low computing cost, low resource occupation, and realize real-time recognition effect Effect

Active Publication Date: 2018-07-20
SHAANXI JUYUN INFORMATION TECH CO LTD
View PDF5 Cites 56 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

For such a large dataset, the traditional Softmax loss requires 8 million output nodes, which will consume a lot of GPU memory

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
  • Face recognition method based on deep neural network
  • Face recognition method based on deep neural network
  • Face recognition method based on deep neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0094] The steps of a face recognition method based on deep neural network include face detection, face alignment, feature extraction and identity comparison;

[0095] Described face detection, the method of aligning are (referring to figure 1 ):

[0096] Use the coarse-to-fine auto-encoding network (CFAN) to detect 5 facial key points (the center of the left and right eyes, the tip of the nose, and the corners of the left and right mouth), and rotate and crop them to 256×256 according to the detected 5 facial key points. By cascading multiple stacked auto-encoder networks, the face alignment results are gradually optimized on face images with higher and higher resolutions;

[0097] The method of feature extraction and identity comparison is as follows:

[0098] A 10-layer depth face network is used to extract face features, and the 10-layer depth face network includes 7 convolutional layers and 3 fully connected layers, which are distinguished by two parts of training and t...

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

The invention relates to a face recognition method based on a deep neural network. A high recognition accuracy rate is realized while the network structure is simplified and the computing time cost islowered. The face recognition method comprises steps of face detection, face alignment, feature extraction and identity comparison. At the face detection and face alignment steps: An automatic codingnetwork (CFAN) from a coarse level to a fine level is used for detecting five facial key points; according to the detected five facial key points, rotary cutting and calibration are carried out to form a forward attitude face image with the pixel of 256*256*3; and on the basis of a plurality of cascaded stack type automatic coding networks, a face alignment result is optimized step by step at a face image with a higher resolution rate. At the feature extraction and identity comparison steps: a face feature is extracted by using a ten-layer deep face network, wherein the ten-layer deep face network consists of seven convolution layers and three fully connected layers; and distinguishing is carried out by training and testing.

Description

technical field [0001] The invention relates to a face recognition method based on a deep neural network. Background technique [0002] Face recognition is a biometric technology for identification based on human facial feature information. A series of related technologies that use a video camera or camera to collect images or video streams containing human faces, automatically detect and track human faces in the images, and then perform facial recognition on the detected faces, usually also called portrait recognition and facial recognition. The traditional face recognition technology is mainly based on face recognition of visible light images, which is also a familiar recognition method and has a research and development history of more than 30 years. However, this method has insurmountable defects, especially when the ambient light changes, the recognition effect will drop sharply, which cannot meet the needs of the actual system. Therefore, robust facial feature repres...

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
IPC IPC(8): G06K9/00G06K9/62G06N3/04
CPCG06N3/04G06V40/165G06V40/168G06V40/172G06F18/22
Inventor 王峰高新波王楠楠
Owner SHAANXI JUYUN INFORMATION TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com