Hair style migration model training method, hair style migration method and related device

A technology for model training and hairstyles, which is applied in the field of image processing, can solve problems such as the blunt fusion effect of virtual hairdressing technology, and achieve the effect of improving the fusion effect and strengthening the characteristics of hairstyles

Pending Publication Date: 2021-07-30
深圳数联天下智能科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] This application provides a hairstyle transfer model training method, a hairstyle transfer method and related devices to solve the problem that the fusion effect of the existing virtual hairdressing technology is relatively blunt

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
  • Hair style migration model training method, hair style migration method and related device
  • Hair style migration model training method, hair style migration method and related device
  • Hair style migration model training method, hair style migration method and related device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] The technical solutions in the embodiments of the present application will be described below with reference to the drawings in the embodiments of the present application.

[0055] The technical solution of this application can be applied to various scenarios of virtual hairdressing, specifically, the technical solution of this application can be applied to various application terminals (such as mobile phones, tablet computers, personal computers, etc.) Hardware (such as smart full-length mirrors, smart hairdressing mirrors, smart cabinets, etc.).

[0056] For ease of understanding, let's first introduce an example of a virtual hairdressing scenario. see figure 1 , figure 1 It is a schematic diagram of a virtual hairdressing scene provided in the embodiment of the application. in, figure 1 A1 in is the user's original hairstyle image, the hairstyle in the original hairstyle image is the user's current real hairstyle, and the user wants to change the hairstyle to fi...

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 provides a hair style migration model training method, a hair style migration method and a related device, wherein the method comprises the steps: obtaining a sample hair style image and a sample reference hair style image; performing feature extraction and feature fusion on the sample hair style image and the sample reference hair style image to obtain a sample fusion feature image; performing feature restoration on the sample fusion feature image to obtain a sample migration hair style image; inputting the sample migration hairstyle image into an image discrimination model and a hairstyle prediction model to obtain a discrimination result and a hairstyle prediction result; and based on the sample migration hairstyle image, the sample hairstyle image, the hairstyle prediction result, the hairstyle in the sample reference hairstyle image and the discrimination result, carrying out iterative parameter adjustment on the hairstyle migration model, the image discrimination model and the hairstyle prediction model until the hairstyle migration model, the image discrimination model and the hairstyle prediction model are converged. By means of the technical scheme, the fusion effect of virtual hairdressing can be more natural.

Description

technical field [0001] The present application relates to the field of image processing, in particular to a hairstyle transfer model training method, a hairstyle transfer method and related devices. Background technique [0002] Beauty salon is a relatively common behavior among users. Before beauty salon, many users want to see their own image after beauty salon. Therefore, some virtual hairdressing applications based on image processing technology emerge as the times require. [0003] In the current virtual hairdressing application, after obtaining the hairdressing image, the hair area is first segmented from the hairdressing image through a segmentation algorithm, and then the template hairstyle in the template is migrated to the hair area, that is, the template hairstyle in the hairdressing template Replace the hairstyle in the hair area, and merge the template hairstyle with other areas in the image to be hairdressed except the hair area to obtain the effect picture aft...

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): G06K9/62G06K9/46G06N3/04G06N3/08
CPCG06N3/08G06V10/462G06N3/045G06F18/253G06F18/214
Inventor 陈仿雄
Owner 深圳数联天下智能科技有限公司
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