Human face image marking method and system

A face image and image technology, applied in the field of face image tagging method and system, can solve the problems of low tagging efficiency, and achieve the effect of high clustering performance, obvious effect, fast tagging and high accuracy

Active Publication Date: 2014-12-31
GUANGZHOU HUADUO NETWORK TECH
View PDF8 Cites 18 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] Based on this, it is necessary to provide a face image tagging method and system for the problem of low tagging efficiency of massive face image data in the prior art.

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
  • Human face image marking method and system
  • Human face image marking method and system
  • Human face image marking method and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0023] see figure 1 as shown, figure 1 For the flow chart of embodiment one of face image labeling method, comprise the following steps:

[0024] Step S101, performing face clustering on the face images to be labeled.

[0025] Step S102, calculating the probability that the face images of each category belong to each labeled person according to the pre-stored classifier model, and labeling the face images according to the probabilities.

[0026] Step S103, training a new classifier according to the labeled face images to update the classifier model.

[0027] The face image labeling method of this embodiment first performs face clustering on the face images that need to be marked, that is, uses a clustering algorithm to cluster the face images into several categories, and if multiple categories belong to the same person, then group them Merge into one category, and the face pictures in each category are of the same person. Then use the face recognition algorithm to identify...

Embodiment 2

[0048] see figure 2 as shown, figure 2 For the flow chart of the second embodiment of the face image labeling method, it comprises the following steps:

[0049] Step S201, put the face images to be marked into the set to be marked.

[0050] Step S202, performing face clustering on the face images in the set to be labeled.

[0051] Step S203, using the preset threshold to judge the number of face images of each category, if it exceeds the threshold, put it into the labeling set, otherwise, keep it in the waiting labeling set.

[0052] Step S204, according to the pre-stored classifier model, calculate the probability that the face images of each category in the labeling set belong to each labeled person, and mark the face images in the labeling set according to the probability, and then put them into the labeled set.

[0053] Step S205, training a new classifier according to the face images in the labeled set to update the classifier model.

[0054] In the face image tagg...

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 human face image marking method includes clustering human face images to be marked; calculating the rate of the varieties of the human face images belonging to marked humans according to pre-stored classifier models, and marking the human face images according to the rate; training fresh classifiers and updating the classifier models according to the marked human face images. The invention further provides a human face image marking system. Thus, fast and accurate human marking of the human face images can be implemented, the marking speed and accuracy rate are increased, the effects of high marking speed and high accuracy are more significant when massive human face images are marked especially, and the massive human face image marking efficiency can be improved greatly.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a face image tagging method and system. Background technique [0002] Face images with marked person information are becoming more and more important. With a large number of marked face images, they can not only be used to train models to improve the effect of face recognition, but also can be used to make many products, such as mobile Internet social networking products etc. [0003] At present, the most common face image labeling method is still manual labeling, through the human recognition ability, manually label each face as a person label. [0004] The speed of manually labeling each picture is very slow. When there are more people to label, the speed of manual recognition will become slower and slower, and the time it takes to label a picture will become longer and longer. When the number of pictures and the number of faces are large, manual annotation is almost...

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): G06K9/66G06K9/00
Inventor 苗广艺路香菊单霆
Owner GUANGZHOU HUADUO NETWORK TECH
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