Fast large-scale face recognition method and system

A face recognition, large-scale technology, applied in the field of face recognition, can solve the problems of slow recognition speed and unstable query performance, and achieve the effect of large index scale, good scalability and strong stability

Active Publication Date: 2022-01-11
湖南视觉伟业智能科技有限公司
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The invention provides a fast large-scale face recognition method and system, which are used to solve the technical problems of slow recognition speed and unstable query performance of existing large-scale face recognition methods

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
  • Fast large-scale face recognition method and system
  • Fast large-scale face recognition method and system
  • Fast large-scale face recognition method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0034] Such as Figure 4 As shown, a fast large-scale face recognition method is disclosed in this implementation, comprising the following steps:

[0035] Collecting a face image to be recognized, and extracting face features from the face image;

[0036] Searching for a sample feature matching the face feature in the face feature cache: if finding a sample feature matching the face feature in the face feature cache, output the identity information of the successfully matched sample feature; If no sample feature matching the face feature is found in the face feature cache, then search for a sample feature matching the face feature in the face feature metric space index library:

[0037] If a sample feature matching the face feature is found in the facial feature metric space index library, the identity information of the successfully matched sample feature is output, and the search time of the face feature is recorded, Judging whether the search time is greater than the pre...

Embodiment 2

[0043] Embodiment 2 is a preferred embodiment of Embodiment 1. It differs from Embodiment 1 in that it introduces the specific structure of the fast large-scale face recognition system, and describes the specific steps of the fast large-scale face recognition method. for refinement:

[0044] Such as figure 1 As shown, in this embodiment, a fast large-scale face recognition system is disclosed, including:

[0045] The video acquisition module collects face video images through the camera;

[0046] Face detection module, which detects faces from the video captured by the camera;

[0047] The feature extraction module is used to extract features from face images, which is realized by a high-precision lightweight neural network model;

[0048] In order to realize the real-time extraction of facial features, the present invention adopts a lightweight deep neural network architecture, which has the remarkable characteristics of high precision and low delay.

[0049] The face fea...

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 discloses a fast and large-scale face recognition method and system. By searching the sample features matching the face features in the face feature cache, if no sample matching the face features is found in the face feature cache features, then search for the sample features that match the face features in the face feature metric space index library: if you find the sample features that match the face features in the face feature metric space index library, then output the matching success The identity information of the sample features, and record the search time of the face features, and judge whether the search time is greater than the preset search time threshold. If it is greater than the search time threshold, add the successfully matched sample features to the face feature cache and use the data survival The time-limited LRU algorithm updates the sample features in the face feature cache. The present invention can improve the efficiency of face feature matching.

Description

technical field [0001] The invention relates to the field of face recognition, in particular to a fast and large-scale face recognition method and system. Background technique [0002] As an important identification technology, face recognition technology has developed rapidly in recent years, and has been widely promoted and applied in social life. Important application fields include transportation, finance, telecommunications, security, education, etc. Although face recognition technology has made significant progress, it still faces many challenges. In particular, large-scale face recognition requires extremely high recognition accuracy. As the scale increases, the comparison of recognition is also extremely time-consuming. How to quickly realize large-scale face recognition in a limited time is a very challenging task. Contents of the invention [0003] The invention provides a fast large-scale face recognition method and system, which are used to solve the technical...

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 Patents(China)
IPC IPC(8): G06V40/16G06V40/50G06V10/94G06F12/123
CPCG06F12/123G06F2212/1024
Inventor 夏东
Owner 湖南视觉伟业智能科技有限公司
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