Face aligning method based on double-layer cascaded neural network

A face alignment and neural network technology, applied in the field of face recognition, can solve problems such as affecting the accuracy of face alignment, and achieve the effects of ensuring performance, reducing detection complexity, and improving detection accuracy.

Active Publication Date: 2018-05-01
SHANGHAI JIAO TONG UNIV
View PDF2 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although this method can achieve face alignment, for all pictures, the initialization of the regressor is obtained by the same prior weak regressor. If an initial shape is too far from the real shape, it is easy to fall into The local optimal solution cannot return to the global optimal solution, which affects the accuracy of face alignment

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 aligning method based on double-layer cascaded neural network
  • Face aligning method based on double-layer cascaded neural network
  • Face aligning method based on double-layer cascaded neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is carried out on the premise of the technical solution of the present invention, and detailed implementation and specific operation process are given, but the protection scope of the present invention is not limited to the following embodiments.

[0037] The invention provides a face alignment method based on a double-layer cascaded neural network. Based on a deep learning algorithm, a convolutional neural network is used to extract features of a face to be detected, and a two-level deep learning model is trained based on samples, and The face alignment problem in the natural environment has been optimized. Such as figure 2 As shown, the method includes the following steps:

[0038] 1) Training a two-level neural network model, the two-level neural network model includes a first-level network for detecting facial contour fea...

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 aligning method based on a double-layer cascaded neural network. The method comprises the following steps that a double layer neural network model which comprises a first-layer network for detecting face contour characteristic points and a five sense organ area and a second-layer network for detecting characteristic points of the five sense organ area, and the fivesense organ area comprises an eyebrow and eye area, a nose area and a mouth area; and 2) and the double layer neural network model carries out 68 characteristic point detection on the picture to be detected, and face alignment is realized. Compared with the prior art, the method has the advantages of high detection precision and high adaption to complex background.

Description

technical field [0001] The invention relates to the technical field of face recognition, in particular to a face alignment method based on a two-layer cascaded neural network. Background technique [0002] Face alignment is one of the hot research topics in computer vision and image fields. Face alignment, also known as facial feature point positioning, is an algorithm that automatically marks the feature points of a face image. The so-called feature points are artificially defined points around the facial features and chin. After face alignment, the facial features of the face can be located, such as figure 1 shown. The reason why face alignment has been widely concerned by researchers is that it is auxiliary to other related technologies such as face matching, face recognition, emotion recognition, etc. In recent years, many excellent algorithms have been used to solve the problem of facial feature point location. However, due to the diversity of facial expressions, pa...

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): G06K9/00G06K9/62G06N3/08
CPCG06N3/08G06V40/161G06F18/253
Inventor 张雨姜飞申瑞民
Owner SHANGHAI JIAO TONG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap