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

Key point and local characteristic-based three-dimensional human face identification method

A three-dimensional face and local feature technology, applied in the field of face recognition, can solve problems such as affecting the recognition effect, indistinguishable recognition algorithms, and loss of face three-dimensional data.

Inactive Publication Date: 2016-11-09
SOUTHEAST UNIV
View PDF0 Cites 27 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] 3D face recognition uses the spatial shape information of the face for identity authentication, and is not affected by factors such as lighting and makeup. Although it has greater advantages compared with 2D face recognition, it still faces the following two challenges: 1. , Expression changes, expressions will cause non-rigid deformation of the face, and it is difficult for the recognition algorithm to distinguish whether the difference between two faces is caused by expression changes or different people
2. Occlusion, occlusion will lead to the loss of three-dimensional data of the face, affecting the recognition effect

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
  • Key point and local characteristic-based three-dimensional human face identification method
  • Key point and local characteristic-based three-dimensional human face identification method
  • Key point and local characteristic-based three-dimensional human face identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0080] The three-dimensional face recognition method based on key points and local features of the present invention realizes the three-dimensional face recognition process through the Matlab R2015b programming tool in the Windows operating system. The experimental data comes from the FRGC v2.0 3D face database, which contains 4007 face models of 466 individuals for testing.

[0081] figure 1 It is an overall flow chart of the inventive method, and the concrete steps are as follows:

[0082] Step 1: Extract the face area from the input 3D face point cloud with the tip of the nose as the center of the sphere and a radius of 90 mm. Perform meshing operation on the extracted 3D face point cloud, use grid-based smoothing algorithm to smooth and denoise the 3D face model, and then restore the smooth 3D face mesh obtained through iterative processing to 3D Face point cloud.

[0083] Step 2: Take the tip of the nose as the center of the sphere and 50mm as the radius to extract the...

Embodiment 2

[0133] Adopt the method of embodiment 1, carry out experimental verification. Specifically include the following steps:

[0134] Step 7: Identity authentication and recognition experiments, all experiments use R1RR (Rank-one Recognition Rate) as the recognition performance index.

[0135] Step 7.1: Experiment 1. This experiment uses the FRGC v2.0 database, which collects 4007 face point clouds of 466 objects, including faces with expressions such as smiling, surprised, and angry. The experiment selects the first One is used as the library set, and the rest are used as the test set, and the recognition rate is 96.9%.

[0136] Step 7.2: Experiment 2. This experiment is based on the Bosphorus database, which collects 4666 face point clouds of 105 objects, in which there are many types of expressions and a large range of expressions. In this experiment, 194 neutral faces are used as the library set, and expressive faces are used as the test set. In the experiment, neutral faces...

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 key point and local characteristic-based three-dimensional human face identification method. The method comprises the steps of preprocessing a three-dimensional human face model by human face region cutting, smoothing processing and pose normalization, and arranging all human faces in a pose coordinate system; detecting key points according to valuable contour lines and an average curvature; constructing a spatial structure of local characteristics in the form of a DAISY descriptor; adopting a shape index histogram, a slant angle histogram and a directional angle histogram as the local characteristics; and performing key point matching, and measuring the similarity of curved surfaces of two human faces by using the number of successfully matched key points. The method is relatively good in identification performance and has the robustness for expression change to a certain extent.

Description

technical field [0001] The invention belongs to the technical field of face recognition, and in particular relates to a three-dimensional face recognition method based on key points and local features, in particular to a face recognition method utilizing key points and local features of the face, especially suitable for facial expressions occasions of change. Background technique [0002] In modern society, more and more people pay more and more attention to accurate and reliable personal identity verification. Identity verification has important applications in many occasions, such as access control systems, video surveillance, and human-computer interaction. Identification technology mainly includes methods based on human biological characteristics such as fingerprints, irises, and faces. Among them, identity authentication based on fingerprints and irises has high accuracy and reliability. And other factors make its application limited, and face recognition has a broader...

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): G06K9/00
CPCG06V20/64G06V40/168
Inventor 达飞鹏郭梦丽汤兰兰邓星何敏成翔昊
Owner SOUTHEAST 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