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

Image face reconstruction method based on encoder-decoder structure

A decoder and encoder technology, applied in the field of deep learning, can solve problems such as high computing resources, poor adaptability to application scenarios, and high real-time requirements, and achieve the effect of reducing computing resource requirements and speeding up face reconstruction

Pending Publication Date: 2022-01-04
际络科技(上海)有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The present invention provides an image face reconstruction method based on an encoder-decoder structure, which is used to solve the adaptability to application scenarios that require high computing resources, high real-time requirements, and limited computing resources in the prior art Poor defects for fast and efficient 3D reconstruction of faces from 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
  • Image face reconstruction method based on encoder-decoder structure
  • Image face reconstruction method based on encoder-decoder structure
  • Image face reconstruction method based on encoder-decoder structure

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. Obviously, the described embodiments are part of the embodiments of the present invention , but not all examples. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0035] Combine below figure 1 , figure 2 The image face reconstruction method based on the encoder-decoder structure of the present invention is described.

[0036] An embodiment of the present invention provides a method for reconstructing an image face based on an encoder-decoder structure, including:

[0037] Step 101, obtaining a face image to be reconstructed;

[0038] Step 103, inp...

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 an image face reconstruction method based on an encoder-decoder structure. The method comprises the following steps: acquiring a face image to be reconstructed; inputting the face image to be reconstructed into an encoder, and obtaining an image feature extraction result based on channel dimension grouping convolution; and inputting the image feature extraction result into a decoder to obtain a UV coordinate matrix as a face reconstruction result. According to the method, the calculated amount is reduced through grouping convolution, the UV coordinate matrix with the adjustable output density is combined, the face reconstruction speed is effectively increased, the requirement for computing resources is reduced, and real-time face three-dimensional reconstruction can be provided through mobile equipment with limited computing resources.

Description

technical field [0001] The invention relates to the technical field of deep learning, in particular to an image face reconstruction method based on an encoder-decoder structure. Background technique [0002] Face 3D reconstruction is a method to obtain accurate face shape, expression, and posture in the driver monitoring system. Traditional 3D face reconstruction relies heavily on the assumption of a priori model of the face. For rare or different faces Without generalization ability, the image face reconstruction algorithm based on deep learning has greatly improved this shortcoming. [0003] However, the image face reconstruction algorithm based on deep learning provided by the existing technology is more focused on high-density accurate modeling. For example, the average face 3D model disclosed in the Basel face database has 53215 vertices and 105842 faces. . Similar reconstruction algorithms require high computing resources, and have poor adaptability to application sc...

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): G06K9/00G06T17/00G06F17/16
CPCG06T17/00G06F17/16
Inventor 宋力程新景
Owner 际络科技(上海)有限公司
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