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

A portrait generation method based on face structure information

A face structure and portrait technology, applied in neural learning methods, biological neural network models, instruments, etc., can solve the problems of face identity information retention, identity information feature extraction and retention, etc.

Active Publication Date: 2021-05-04
HANGZHOU DIANZI UNIV
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the extraction and retention of identity information features is still a difficult problem in the actual sketch generation, especially in the pursuit of visual effects, the problem of face identity information retention will be more difficult

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 portrait generation method based on face structure information
  • A portrait generation method based on face structure information
  • A portrait generation method based on face structure information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0064] The detailed parameters of the present invention will be further specifically described below.

[0065] Such as figure 1 As shown, a portrait generation method based on face structure information includes the following steps:

[0066] Step (1), data preprocessing:

[0067] Face alignment is performed on face photos and portraits with the original size of 250×200. The image size of the aligned face photos is 250×200, the distance between the eyes is 50, and the distance from the eyes to the top of the image is 125; Fill the edge of the image to get a 286×286 image, randomly select a face photo image X of size 256×256 each time for training; use the U-Net network to extract features with spatial and texture information;

[0068] The aligned face photo image X corresponds to the sketch image Y of the face photo and the probability map of the face structural components Form the triplet, X, Y, as a training set;

[0069] Step (2), based on the probability map of face ...

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 portrait generation method based on human face structure information. The invention includes the following steps: 1. Data preprocessing is performed on the original image, the target image and the face structure information, and 2, the face structure information model is used at the input end of the image generator to perform feature extraction and fusion. 3. Use the combined loss function based on the structural components of the face in the loss function part of the image generator. 4. Generative adversarial network, use the generator to generate, and use the discriminator to differentiate. 5. Model training, using back-propagation algorithm to train neural network parameters. The present invention proposes a neural network model for generating from a face photo to a portrait, and particularly proposes that the portrait generator uses the instructional information of the face parts to generate the portrait and uses the part information to calculate the loss of each part. A method for network parameter optimization, which achieves the best results in the field of face photo-to-portrait generation.

Description

technical field [0001] The present invention relates to a generation confrontation network generated from a face photo to a portrait (Photo-Sketch Synthesis), which mainly relates to a method for modeling from a face photo-portrait generation, and using face structure information to image The generation of the guided optimization. Background technique [0002] Face sketch portrait generation (Sketch Portrait Generation) problem is to generate a corresponding face sketch given a face, and it is also called photo-sketch transformation (Photo-sketch Transformation) or face sketch generation (Face Sketch Synthesis). Face sketch generation has many applications, such as entertainment or criminal investigation. An ideal generated face sketch should have two characteristics, one is to retain the appearance of the person, so that there is a high readiness rate in the recognition of the identity information of the sketch face sketch; the other is to be like a sketch, so that it It...

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): G06K9/00G06N3/08
CPCG06N3/084G06V40/171
Inventor 俞俊施圣洁高飞
Owner HANGZHOU DIANZI 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