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

Semi-supervised face age estimation device and semi-supervised face age estimation method

A semi-supervised, face-based technology, applied in computing, computer components, instruments, etc., can solve the problem of age-marked face pictures in face pictures, and achieve good age estimation accuracy

Active Publication Date: 2019-01-18
SOUTHEAST UNIV
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The above method can estimate the age of face pictures more accurately, but the face pictures used in its training can only be face pictures with age marks

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
  • Semi-supervised face age estimation device and semi-supervised face age estimation method
  • Semi-supervised face age estimation device and semi-supervised face age estimation method
  • Semi-supervised face age estimation device and semi-supervised face age estimation method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0019] The technical scheme of the present invention is described in detail below in conjunction with accompanying drawing:

[0020] The semi-supervised human face age estimation method of the present invention includes two stages: the training stage of the estimation model and the age estimation stage. Among them, the estimation model is trained as figure 2 As shown, it specifically includes the following steps:

[0021] Step 1, obtain face picture data set, and carry out image feature extraction to each face picture wherein; Described face picture data set comprises a group of age-marked face pictures and a group of age-free face pictures .

[0022] The face picture data set used for model training in the present invention includes both face pictures with age marks and face pictures without age marks, so that the scale of the face picture data set can be greatly expanded, and a large number of easy-to-obtain unlimited data sets can be fully utilized. Age-tagged images of...

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 semi-supervised human face age estimation device. The training method of the device is as follows: step 1, obtain a face picture data set, and perform image feature extraction; step 2, for each face picture with an age mark, initialize the age distribution for it, and use it as a training set; step 3 1. Utilize the current training set to train the LBFGS-LLD model, and predict the age distribution of all pictures; step 4, calculate the pseudo-age of face pictures without age marks; Corresponding variance, and use the obtained variance to update the age distribution of face pictures in the corresponding age group; step 6, use all the updated pictures as a new training set, go to step 3; until the iteration termination condition is met. The invention also discloses a semi-supervised human face age estimation method based on the device. The present invention only needs to use a small number of age-marked face pictures combined with more age-free face pictures to obtain better age estimation accuracy.

Description

technical field [0001] The invention relates to a human face age estimation device, in particular to a semi-supervised human face age estimation device and a semi-supervised human face age estimation method, which belong to the technical field of machine learning and pattern recognition. Background technique [0002] Age is one of the important attributes of people, and people's behavior and preferences are different in different age groups, which indicates that accurate age estimation will have a very important application prospect. At present, many age-related applications, such as human-computer interaction, electronic customer relationship management, security management, monitoring and monitoring, etc., have been greatly developed. Among the many age estimation methods, face-based age estimation may be the most commonly used in daily life. [0003] A Chinese invention patent (CN102567719) discloses a "human age automatic estimation method based on the posterior probabi...

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): G06K9/00G06K9/62
CPCG06V40/161G06V40/178G06V40/172G06F18/2413
Inventor 耿新侯鹏
Owner SOUTHEAST UNIV
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