A method for predict face change based on generate antagonistic network

A prediction method and network technology, applied in the field of domain adaptation and face recognition, can solve the problems of facial expressions and even identity information deviation, without emphasizing identity information, etc., to ensure uniqueness, improve accuracy and uniqueness, and improve robustness. Effect

Inactive Publication Date: 2019-02-05
NEXWISE INTELLIGENCE CHINA LTD +1
View PDF2 Cites 41 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At the same time, this technology mainly considers the probability distribution of the face set without emphasizing identity information, resulting in large deviations in face color, expression and even identity information when performing face age conversion

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 method for predict face change based on generate antagonistic network
  • A method for predict face change based on generate antagonistic network
  • A method for predict face change based on generate antagonistic network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings. It is only stated here that the words for directions such as up, down, left, right, front, back, inside, and outside that appear or will appear in the text of the present invention are only based on the accompanying drawings of the present invention, and are not specific to the present invention. limited.

[0047] The main purpose of the present invention is to predict the transformation of face images of different age groups based on existing face images, and to generate images close to the real situation as much as possible under the premise of ensuring the uniqueness of identity. In order to predict the future changes of face images more accurately, the present invention proposes a new face change prediction method based on Generative Adversarial Networks (GAN, G...

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 face change prediction method based on a generation antagonism network, which comprises the following steps: S1, acquiring a face data sample; S2, constructing a generation antagonism network model composed of a generator and a discriminator, designing a loss function, and iteratively training the generator and the discriminator; S3, inputting the target young person faceimage to be processed into the generated antagonism network model after training, and outputting the target old person face image corresponding to the target young person face image. Adopting the invention, by constructing a unique generation antagonism network model and a loss function, the invention has good robustness to age change, and at the same time, the invention also considers the information changes of forehead and hair, and further improves the accuracy and uniqueness of prediction.

Description

technical field [0001] The present invention relates to the technical fields of domain adaptation and face recognition, in particular to a face change prediction method based on a generative confrontation network. Background technique [0002] Facial change prediction mainly refers to predicting changes in the face with age. It is usually applicable to the entertainment industry or important evidence collection sites. For example, the public security system predicts the current appearance of existing files based on previous images. [0003] Due to the inherent complexity of human body age and the lack of labeled data, facial change prediction is very difficult. Usually when solving such problems, the accuracy of predicting age and identity uniqueness are regarded as two important indicators. The common method in previous years is to infer the changes of facial muscles over time based on the structural characteristics of the skin, which is more complicated in practice. [0...

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): G06K9/00
CPCG06V40/161G06V40/168
Inventor 胡建国王金鹏王德明吴明华杨焕马媛
Owner NEXWISE INTELLIGENCE CHINA LTD
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