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

Lightweight expression recognition method based on deep learning

An expression recognition and deep learning technology, applied in neural learning methods, character and pattern recognition, acquisition/recognition of facial features, etc., can solve problems such as poor real-time performance and large recognition model, to prevent overfitting and achieve good robustness. , the effect of high operating efficiency

Inactive Publication Date: 2020-12-18
SOUTHWEAT UNIV OF SCI & TECH +1
View PDF4 Cites 22 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is to provide a light-weight expression recognition method based on deep learning to solve the problems of large and poor real-time performance of the existing deep learning-based expression recognition model proposed in the above background technology

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
  • Lightweight expression recognition method based on deep learning
  • Lightweight expression recognition method based on deep learning
  • Lightweight expression recognition method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] In order to enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be described in detail below in conjunction with the accompanying drawings. The description in this part is only exemplary and explanatory, and should not have any limiting effect on the protection scope of the present invention. .

[0027] Such as Figure 1-Figure 5 As shown, the concrete structure of the present invention is: a kind of light-weight facial expression recognition method based on deep learning, comprises the following steps,

[0028] S10. MTCNN face detection and positioning. After obtaining the image collected by the camera or inputting a picture, first use the small model to generate a certain possibility of face candidate frames, and then use a slightly more complex network to screen and classify the candidate frames Regression with a higher-precision area frame, and let this step be performed recursively, and u...

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 the related technical field of lightweight expression recognition methods, in particular to a deep learning-based lightweight expression recognition method, which comprises the following steps that firstly, quick and efficient face detection is realized through an MTCNN lightweight face detection network; then, the basic structure of the MobileNetV3 network is improved, and a linear bottle is improved on the basis of depth separable convolution, so that the network complexity is reduced, and over-fitting is prevented; and finally, an efficient attention module is designed to combine the depth and spatial information of the feature map, important feature extraction is more emphasized, an intra-class feature difference of the same expression is reduced by adopting aCenter loss function, and Softmaxloss expands a feature distance between different expression classes, so that the network has a better feature discrimination effect, and finally, real-time facial expression recognition is completed.

Description

technical field [0001] The present invention relates to the related technical field of lightweight facial expression recognition methods, in particular to a lightweight facial expression recognition method based on deep learning. Background technique [0002] Facial expression is a common form of non-verbal communication, which can effectively convey personal emotions and intentions; humans can obtain the facial expressions of others through vision and understand the inner state of others through brain analysis to achieve the purpose of communication. With the development of prosperity and artificial intelligence, people hope that machines can recognize facial expressions relatively accurately to achieve communication between humans and machines. Automatic facial expression recognition is improving human-computer interaction, distance education, auxiliary medical care, driving fatigue monitoring, marketing assistance, etc. All aspects have important research value and wide a...

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): G06K9/00G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06V40/174G06V40/172G06N3/048G06N3/045G06F18/241
Inventor 张红英韩兴吴亚东
Owner SOUTHWEAT UNIV OF SCI & TECH
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