A face image restoration method based on a generation antagonism network

A face image and repair method technology, applied in the field of deep learning and image processing, can solve problems such as model collapse, low similarity, unstable network training, etc., to achieve the effect of improving stability and solving instability

Active Publication Date: 2019-02-22
BEIJING UNIV OF TECH
View PDF3 Cites 72 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] There are two technical problems to be solved by the present invention. One is that the existing generative confrontation network has problems of unstable network training and mode collapse; the other is that the existing face repair images do not conform to visual cognition and the similarity is not high.
Aiming at these two problems, the present invention proposes a network design scheme that can not only solve the problems of unstable training and mode collapse of the generative confrontation network, but also generate and complement more natural and realistic face images

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 image restoration method based on a generation antagonism network
  • A face image restoration method based on a generation antagonism network
  • A face image restoration method based on a generation antagonism network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0063] Method of the present invention comprises the following steps:

[0064] Step 1, face data preprocessing stage. Set the size of the collected data to obtain the size of the face required for training.

[0065] Perform face recognition on the collected images, extract face information, the top of the chin, the outer edge of the eyes, the inner edge of the eyebrows, etc.; according to the landmarks located on each face, the collected images are cropped into a set Face training images sized so that the eyes and mouth are centered

[0066] Step 2, the training phase. Let the processed face data set be used as training data to train the generative confrontation network.

[0067] GAN consists of two networks: the generation network G and the discriminative network D, the structure is as follows figure 1 As shown, the purpose of the generation network G is to generate face images similar to the real data distribution, and the purpose of the discriminant network D is to judg...

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 image restoration method based on a generation antagonism network. The method comprises the following steps: a face data set is preprocessed, and a face image with a specific size is obtained by face recognition of the collected image; In the training phase, the collected face images are used as dataset to train the generating network and discriminant network, aimingat obtaining more realistic images through the generating network. In order to solve the problems of instability of training and mode collapse in the network, the least square loss is used as the loss function of discriminant network. In the repairing phase, a special mask is automatically added to the original image to simulate the real missing area, and the masked face image is input into the optimized depth convolution to generate an antagonistic network. The relevant random parameters are obtained through context loss and two antagonistic losses, and the repairing information is obtainedthrough the generated network. The invention can not only solve the face image repairing with serious defective information, but also generate a face repairing image which is more consistent with visual cognition.

Description

technical field [0001] The invention belongs to the field of deep learning and image processing, and in particular relates to a face image restoration method based on a generative confrontation network. Background technique [0002] Image inpainting technology is an important branch in the field of image processing in recent years, which belongs to the interdisciplinary problems of pattern recognition, machine learning, statistics, computer vision and so on. Image restoration refers to the restoration and reconstruction of the missing image information caused by the image preservation process or the restoration after removing redundant objects in the image. Nowadays, researchers have proposed a variety of image restoration methods, which have been widely used in the fields of old photo restoration, cultural relic protection, and removal of redundant objects. [0003] Due to the inherent fuzziness and complexity of natural images, the traditional methods based on texture and...

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): G06T5/00
CPCG06T5/005G06T2207/20081G06T2207/30201
Inventor 任坤孟丽莎杨玉清
Owner BEIJING UNIV OF TECH
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