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

Surface normal component graph neural network representation-based three-dimensional face recognition method

A 3D face and neural network technology, applied in the field of 3D face recognition, can solve problems such as difficult artificial features

Pending Publication Date: 2018-03-27
XI AN JIAOTONG UNIV
View PDF2 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, for a specific task, it is not easy to design corresponding artificial features

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
  • Surface normal component graph neural network representation-based three-dimensional face recognition method
  • Surface normal component graph neural network representation-based three-dimensional face recognition method
  • Surface normal component graph neural network representation-based three-dimensional face recognition method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0053] The present invention will be described in detail below in conjunction with the accompanying drawings.

[0054] see figure 2 , the present invention comprises the following steps:

[0055] Step 1. Given a 3D face surface, preprocessing is performed first; the preprocessing includes: denoising of the 3D face surface, automatic detection of nose point, face area cutting and pose registration.

[0056] The specific process of preprocessing is as follows: first, perform Gaussian filtering and median filtering on the original 3D face surface, calculate the Gaussian curvature and median curvature on the filtered face surface, and determine the shape of the tip of the nose according to the curvature information combined with a 3D model. Position; the face area is defined as the intersection of the 90cm radius sphere and the three-dimensional face surface with the tip of the nose as the center of the sphere. Use the classic ICP (Iterative Closet Point) algorithm for pose reg...

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 surface normal component graph neural network representation-based three-dimensional face recognition method. The method includes the following steps that: a given three-dimensional face surface is preprocessed; the three-dimensional coordinates of the pre-processed three-dimensional face surface are projected onto a two-dimensional plane, projected coordinate information is utilized to estimate the normal vectors of each point, so that normal component graphs in an X, Y, and Z direction are generated; a deep convolutional neural network trained in 2D face recognition is utilized to extract the features of each normal component graph of the three-dimensional face; and the depth features of each normal component graph are extracted under a nerve cell position-sensitive matching mode, and a nearest neighbor classifier or a sparse representation classifier is used to achieve the comparison of the three-dimensional face. The three-dimensional face recognition technology of the present invention has the advantages of simplicity, easiness in implementation, high robustness to expressions, high recognition accuracy and the like.

Description

technical field [0001] The invention relates to a three-dimensional human face recognition method, in particular to a three-dimensional human face recognition method based on a surface normal component graph neural network representation. Background technique [0002] As a new type of biometric identification technology, 3D face recognition technology has great potential application value in finance, security, anti-terrorism and other fields. The core of 3D face recognition technology is to accurately describe the shape of the 3D face surface. The existing technology mainly includes the characterization of geometric quantities such as points, lines, surfaces, normal vectors, curvatures, and shape indicators based on three-dimensional human face surfaces. At the same time, it combines artificially designed features (such as Gabor wavelet transform and local binary mode) to achieve the final expression of the three-dimensional face surface. In particular, existing related te...

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): G06K9/00G06K9/46
CPCG06V40/168G06V40/172G06V10/40G06V10/513
Inventor 李慧斌孙剑魏晓帆
Owner XI AN JIAOTONG 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