Self-adaptive feature fusion recovery and mixed makeup migration recombination method

A feature fusion and self-adaptive technology, applied in neural learning methods, instruments, biological neural network models, etc., can solve the problem that details are not as good as using variational autoencoder to generate network technology, achieve good makeup transfer effect, improve lu Stickiness, the effect of improving the migration effect

Pending Publication Date: 2022-07-29
SHAANXI NORMAL UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Experiments show that using the generative confrontation network for global makeup transfer is better, but the local details are not as good as using the variational autoencoder to generate the network technology

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
  • Self-adaptive feature fusion recovery and mixed makeup migration recombination method
  • Self-adaptive feature fusion recovery and mixed makeup migration recombination method
  • Self-adaptive feature fusion recovery and mixed makeup migration recombination method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0075] The present invention will be further described below in conjunction with specific embodiments. It should be understood that these examples are only used to illustrate the present invention and not to limit the scope of the present invention. In addition, it should be understood that after reading the content taught by the present invention, those skilled in the art can make various changes or modifications to the present invention, and these equivalent forms also fall within the scope defined by the appended claims of the present application.

[0076] In order to prevent the influence of the clarity of the input image and the existing background or the difference in the pose and expression of the characters on the makeup migration effect, the present invention provides a facial feature fusion and restoration implementation method and a guide image that can achieve a relatively optimal makeup migration effect. The implementation of the specific selection includes the fo...

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 relates to a self-adaptive feature fusion recovery and mixed makeup migration recombination method. According to the method, a facial feature recovery technology based on an adaptive feature fusion module is mainly realized, a guide image is selected according to a solved weighted least square WLS model, and a feature extraction network and a style migration network are established to perform makeup removal operation; and realizing makeup migration and recombination based on the reference makeup image by combining respective advantages of the generative adversarial network and the variational auto-encoder network. The method mainly solves the problem that due to different postures or illumination, difference of backgrounds and figure postures, definition and size of input photos and built-in makeup, the makeup migration and recombination effect is affected. And makeup reorganization is carried out by combining a generative adversarial network and a variational auto-encoder network technology, so that a better makeup migration effect is achieved.

Description

technical field [0001] The invention relates to the field of computer image processing, in particular to a facial feature fusion recovery technology based on an attention mechanism of adaptive spatial feature fusion and a makeup reorganization method combining generative confrontation network and variational autoencoder generation network technology. Background technique [0002] In the field of computer image processing, makeup transfer is a branch of style transfer, and the use of generative adversarial network can better realize the separation of makeup features and face features from the source image and the makeup reorganization of makeup transfer on the target image. However, it is easily affected by the differences in background and character poses between the reference image and the source image. In order to optimize the effect of makeup transfer, from paired makeup transfer to unpaired makeup transfer by introducing a cyclic sustainable loss function. [0003] In t...

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): G06T3/00G06K9/62G06N3/04G06N3/08G06V10/80G06V10/774G06V10/82
CPCG06N3/084G06N3/045G06F18/253G06T3/04
Inventor 李宇苗李鹏
Owner SHAANXI NORMAL UNIV
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