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

Face super-resolution reconstruction method, reconstruction device and computer system

A super-resolution reconstruction and high-resolution technology, applied in the field of face super-resolution reconstruction, reconstruction equipment and computer systems, can solve problems such as failure to consider structured information

Inactive Publication Date: 2018-06-08
BOE TECH GRP CO LTD
View PDF3 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] None of the above takes structured information into account

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
  • Face super-resolution reconstruction method, reconstruction device and computer system
  • Face super-resolution reconstruction method, reconstruction device and computer system
  • Face super-resolution reconstruction method, reconstruction device and computer system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0061] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some of the embodiments of the present invention, but not all of them.

[0062] The present invention proposes a face super-resolution reconstruction method based on machine learning. The method is based on a deep learning (deep learning) convolutional neural network intelligent algorithm to construct a standard high-resolution gradient image library. On this basis, mark the feature points and pose estimation of the input low-resolution image; further obtain the corresponding high-resolution facial components and similar edge prior information; extract and integrate the corresponding gradient information of each part; superimpose to the original From the low-resolution image, a high-resolution image of the face is obtained. T...

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 face super-resolution reconstruction method based on machine learning. According to the method, a high-resolution image sample set is constructed, and low-resolution imagesare inputted. The method comprises the following steps of 1.1, determining the edge gradient of an input low-resolution image; 1.2, based on the high-resolution image sample set, determining a high-resolution feature gradient corresponding to the input low-resolution image; 1.3, fusing the edge gradient and the high-resolution feature gradient, superposing the fused gradient information into the input low-resolution image to obtain a super-resolution image with the resolution thereof higher than that of the input low-resolution image. The invention further relates to a corresponding reconstruction device.

Description

technical field [0001] The present invention relates to a method for super-resolution reconstruction of human face based on machine learning and a corresponding reconstruction device and computer system constructed to perform the reconstruction method. Background technique [0002] The face image super-resolution reconstruction technology is mainly suitable for enlarging the photos stored in the existing IC card for easy viewing and printing. Super-resolution reconstruction techniques are especially applicable in many situations where the cost of replacing existing (storage. acquisition) equipment is high and re-acquisition is low. [0003] Or perform super-resolution processing on face images acquired from surveillance equipment for easy identification. Due to hardware technology, cost and other limitations, it may not be possible to collect clear high-resolution images in the monitoring field. The use of super-resolution reconstruction technology can reduce the dependence...

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): G06T3/40
CPCG06T3/4053G06T3/4076G06T2207/30201G06T2207/20221G06T5/50
Inventor 张丽杰
Owner BOE TECH GRP CO LTD