Multi-modal human face recognition method based on deep learning

A face recognition and deep learning technology, applied in the field of face recognition, can solve problems such as low efficiency and insufficient expression ability, and achieve the effect of improving performance and fast and accurate face recognition

Active Publication Date: 2017-06-30
SEETATECH BEIJING TECH CO LTD
View PDF6 Cites 58 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the features used in this invention are all artificially designed features, and the expressive ability is not strong enough; and the features of this invention use Adaboost algorithm for feature fusion and feature selection, which is inefficient; and this invention is designed for two specific modes , with limitations

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
  • Multi-modal human face recognition method based on deep learning
  • Multi-modal human face recognition method based on deep learning
  • Multi-modal human face recognition method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0022] Such as figure 1 Shown, the present invention specifically comprises the following steps:

[0023] (1) Perform face detection, feature point location, alignment, and cropping on the RGB face image, and make the cropped RGB mode face data set S0; then according to the RGB mode and other modes (such as depth information, Near-infrared information, etc.), find the feature points of other modal faces, and cut and make other modal face datasets S1, S2...;

[0024] (2) Design a multimodal fusion deep neural network structure N1. In this structure, the first half is several independent neural network branches, and the input of each branch corresponds to a modality (such as RGB modality, deep modality , NIR mode, etc.), and then use a specific network structure to fuse multiple modal branches into a synthetic neural network branch (such as connecting these features, or stacking them by channels, or other connection structures, such as attaching image 3 Such a structure, etc....

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 multi-modal human face recognition method based on deep learning, and the method comprises the steps: (1), carrying out the human face detection and alignment of an RGB human face image; (2), designing a multi-modal fusion depth convolution neural network structure N1, and training an N1 network; (2), designing a multi-modal shared depth convolution neural network structure N2, and a training an N2 network; (4), extracting features; (5), calculating similarity; and (6), carrying out the similarity fusion. The method employs a multi-modal system, overcomes some inherent shortcomings of a single-modal system through the collection of various types of human face modal data, the advantages of various types of modal information and a fusion strategy, effectively improves the performance of a human face recognition system through the efficient use of various types of modal information, and enables the human face recognition to be quicker and more accurate.

Description

technical field [0001] The present invention relates to a face recognition method, in particular to a multi-modal face recognition method based on deep learning. Background technique [0002] Compared with 2D face recognition, 3D face recognition has the advantages of being robust to illumination and less affected by factors such as posture and expression. Therefore, after the rapid development of 3D data acquisition technology and the great improvement in the quality and accuracy of 3D data, Many scholars devote their research to this field. [0003] The images of different modalities of the face are easily affected by different factors, etc. These factors affect the stability and accuracy of the single-modal face recognition system to a certain extent. CN104778441A proposes a multimodal face recognition device and method that fuses grayscale information and depth information, and its core method is after extracting multimodal facial features (the features used in the inve...

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): G06K9/00G06K9/62
CPCG06V40/161G06F18/25
Inventor 张浩韩琥山世光陈熙霖
Owner SEETATECH BEIJING TECH CO LTD
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