A method for recognizing facial expressions

A face expression and recognition method technology, applied in the field of face recognition, can solve the problem that the real-time computing system cannot support the calculation amount of the algorithm, and achieve the effect of improving the description ability, the accuracy rate and the expression ability.

Inactive Publication Date: 2020-08-11
BEIJING NORMAL UNIVERSITY
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the amount of existing sample data is difficult to meet the requirements of this type of algorithm, and real-time computing systems, especially mobile devices, cannot support the amount of computing required by the algorithm.

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
  • A method for recognizing facial expressions
  • A method for recognizing facial expressions
  • A method for recognizing facial expressions

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] The specific implementation of the identification method of the present invention will be further described in detail below in conjunction with the drawings in the description.

[0046] like figure 1 and figure 2 As shown, the identification method of the present invention specifically includes the following steps:

[0047] Step 1: For each image in the facial expression dataset, detect the face and feature points in the image:

[0048] Step 1.1 For each image in the facial expression training data set, detect the face in the image and realize the feature point alignment. In order to accurately detect the exact position of different face feature points, it is necessary to use the manually marked feature points in the sample face to average The position of the face feature points is optimized for regression calculation, and the optimization function is defined as the following formula (1):

[0049]

[0050] in Represents the feature points of the average human f...

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 present invention relates to a kind of recognition method of human facial expression, specifically comprises the following steps: step 1: for each image in the human facial expression data set, detect the human face and feature point in the image; step 2: on the basis of step 1, Generate the feature vectors of the local area of ​​each feature point on the face image, and calculate the expression features based on the face feature points; Step 3: Calculate the face expression feature vector; Step 4: On the basis of Step 3, use the self-encoding neural network Method for dimensionality reduction; Step 5: Calculate and obtain a nonlinear high-dimensional classification model; Step 6: After the user inputs the face video, construct a new lower-dimensional shape feature descriptor according to Step 1 to Step 4; Step 7: The feature descriptor of the input video face image is compared with the classification model to determine the probability value of the facial expression image in different classifications. The invention has the beneficial effects of: improving the description ability of expression features; improving the accuracy of expression recognition.

Description

technical field [0001] The invention relates to the technical field of face recognition, in particular to a method for recognizing facial expressions. Background technique [0002] Facial expressions are expressed through the shape changes of different parts of the face, and then express various emotional states and convey different emotions. It has a very broad application prospect to analyze and recognize facial expressions by using image geometric shape changes. [0003] Facial expression recognition can be applied to personalized customization, judge the emotional state according to the user's facial expression recognition, and then recommend various services that meet the user's needs, including entertainment content recommendation, life service information recommendation, etc.; can be applied to online learning, Judging the user's interest in the learning content according to the expression change, automatically evaluating and adjusting the learning content, and guidi...

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 Patents(China)
IPC IPC(8): G06K9/00
CPCG06V40/171G06V40/174
Inventor 樊亚春税午阳宋毅
Owner BEIJING NORMAL UNIVERSITY
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