Unlock instant, AI-driven research and patent intelligence for your innovation.

Super-resolution face image reconstruction method

A face image and super-resolution technology, which is applied in the field of face images, can solve the problems of high complexity of reconstructing super-resolution face images and low efficiency of super-resolution face images, so as to reduce the complexity and the amount of computation , simple structure and reduced complexity

Active Publication Date: 2021-07-30
SOUTHWEST JIAOTONG UNIV
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to solve the technical problem that the complexity of reconstructing a super-resolution face image in the prior art is too high, and the efficiency is low when reconstructing a super-resolution face image, and proposes a super-resolution face image reconstruction method

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
  • Super-resolution face image reconstruction method
  • Super-resolution face image reconstruction method
  • Super-resolution face image reconstruction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0043] Next, the technical scheme in the present application embodiment will be described in the present application, and it is understood that the described embodiments are intended to be described herein, not all of the embodiments of the present application. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative labor, are the scope of the present application.

[0044] As described in the context, the prior art research on human face super resolution is mainly a single-resolution study, which is because it is difficult to get a multi-frame image of the same face in practical applications, and nearly a few With the development of deep convolutional neural network, the super-resolution study of single-image tend to use complex network simulation mapping relationships from low-resolution images to high-resolution images, such technical solutions are gaining higher reconstruction quality. At the same time,...

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 super-resolution face image reconstruction method, and the method comprises the steps: obtaining a face image training set, building a face image network, inputting the face image training set into the face image network, and obtaining a super-resolution training face image set; calculating the loss function value based on the super-resolution training face image set and the face image training set, training the face image network by optimizing the loss function, and performing super-resolution face image reconstruction based on the low-resolution face image and the trained face image network, so that the complexity of super-resolution face image reconstruction is reduced, the fast and accurate reconstruction of the super-resolution face image is realized, and the reconstructed super-resolution face image is a high-quality face image.

Description

Technical field [0001] The present invention belongs to the field of human face image, and specific relates to a super-resolution human face image reconstruction method. Background technique [0002] Human face super resolution technology is a special super-resolution technology that can utilize a low-resolution human face image to obtain a high-resolution human face image with real high-frequency detail, but also in the field of video security and film and television entertainment. With a wide range of applications, it can also help other face-related tasks have better results. [0003] The research on human face super resolution is primarily for single-resolution, and with the development of deep convolutional neural networks, the study of single-resolution is trend to use complex network simulation by lower resolution. The image to the mapping relationship of high-resolution images, increasing the complexity and training difficulty of the network. [0004] Therefore, how to re...

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
IPC IPC(8): G06T3/40G06N3/04G06N3/08G06K9/62
CPCG06T3/4053G06N3/084G06T2207/30201G06T2207/20081G06N3/045G06F18/214
Inventor 和红杰蒋桐雨陈帆
Owner SOUTHWEST JIAOTONG UNIV