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

A face multi-attribute recognition method based on multi-source data

A technology of multi-source data and recognition methods, applied in the field of face attribute recognition, can solve problems such as hindering research progress, unable to provide all attribute labels, and unbalanced distribution of data set categories. Ability to optimize and reduce the effect of overfitting

Active Publication Date: 2022-08-05
UNIV OF ELECTRONICS SCI & TECH OF CHINA
View PDF12 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In addition, with the continuous expansion of the demand for face attribute research, some problems have been exposed in the current face attribute data set, which seriously hinders the research progress in this direction.
For example, the category distribution of the data set is unbalanced, or for multi-attribute recognition tasks, a single data set cannot provide all attribute labels, etc.

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
  • A face multi-attribute recognition method based on multi-source data
  • A face multi-attribute recognition method based on multi-source data
  • A face multi-attribute recognition method based on multi-source data

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] The technical solutions of the present invention have been described in detail in the section of the content of the invention. In order to make those skilled in the art better understand the method of the present invention, supplementary descriptions are given below with reference to the accompanying drawings.

[0029] like figure 1 As shown, it is the overall implementation process of the present invention, wherein, for the input face image, a preprocessing process is included, and the specific method is:

[0030] 1) Perform face key point detection on the input image;

[0031] 2) Align the face according to the key points of the face;

[0032] 3) Crop the aligned face, remove useless information, and retain only the part of the face that contains valid information;

[0033] 4) Scale the cropped face image to a fixed size corresponding to the input of the model.

[0034] figure 2 The face attribute recognition network model proposed by the present invention includ...

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 belongs to the technical field of face attribute recognition, in particular to a face multi-attribute recognition method based on multi-source data. Aiming at the problem that a single face attribute data set often contains limited attribute labels and often cannot meet the labeling requirements for all attributes in the multi-attribute recognition task, the invention proposes a face multi-attribute recognition method based on multi-source data. The specific requirements of the face attribute recognition task can be combined with multiple existing data sets to meet the multiple attribute labeling requirements required by the task, and through joint training, multiple data sets can promote each other, and the face The attribute recognition network is effectively trained, so as to realize the joint recognition of various attributes of the input face image. The beneficial effect of the invention is that, while satisfying all attribute labeling requirements of the task, the information interaction between multiple attributes is realized through the mutual cooperation of multiple data sets, the joint training objectives of multiple attributes are completed, and the network generalization effect is improved.

Description

technical field [0001] The invention belongs to the technical field of face attribute recognition, in particular to a face multi-attribute recognition method based on multi-source data. Background technique [0002] With the rapid rise of artificial intelligence and the continuous development of image acquisition and processing technology, artificial intelligence applications for information mining and utilization of images can be seen everywhere in people's lives. As an important biological feature, human face contains a lot of important information, and it is more and more favored by researchers because of its rich details and easy access. [0003] The initial research on face data is face recognition and face detection, which is the most direct application of face information. With the continuous growth of people's material and cultural needs, researchers are also constantly developing the use of face information, and face attribute recognition has emerged as the times r...

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): G06V40/16G06V10/774G06V10/764G06V10/766G06K9/62G06V10/82G06N3/04
CPCG06V40/172G06N3/045G06F18/2148G06F18/24
Inventor 朱策胡佃敏章超张铁刘翼鹏
Owner UNIV OF ELECTRONICS SCI & TECH OF CHINA
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