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

A cross-dataset facial expression recognition model construction and recognition method

A facial expression, recognition model technology, applied in character and pattern recognition, acquisition/recognition of facial features, computer parts and other directions, can solve problems such as unsatisfactory clustering effect

Active Publication Date: 2019-01-04
CHANGAN UNIV
View PDF2 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a cross-dataset facial expression recognition model construction and recognition method to solve the problem of unsatisfactory clustering effect of cross-dataset expression recognition methods in the prior art

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 cross-dataset facial expression recognition model construction and recognition method
  • A cross-dataset facial expression recognition model construction and recognition method
  • A cross-dataset facial expression recognition model construction and recognition method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] The invention discloses a method for constructing a facial expression recognition model across data sets, the method comprising:

[0032] Step 1, obtain a training set of facial expression images, the training set includes a plurality of marked facial expression images in the first data set and a plurality of unlabeled facial expression images in the second data set, in the training set Each labeled facial expression image corresponds to a labeled label, and each unlabeled facial expression image corresponds to a random label;

[0033] In this embodiment, 327 facial expression images of marked expressions in the CK+ data set and 142 facial expression images of unmarked expressions in the JAFFE data set are respectively selected to form a set of facial expression image training sets, wherein the facial expression images in the CK+ data set are such as figure 2 As shown, the facial expression images in the JAFFE dataset are as follows image 3 shown. The labels of the ...

Embodiment 2

[0080] In a facial expression recognition method across data sets, the facial expression recognition model described in the first embodiment recognizes the facial expression images to be recognized, and obtains the labels of the facial expression images to be recognized.

[0081] In this embodiment, for Figure 4 The facial expression image shown is recognized, and the recognition result is: 7-happy.

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 method for constructing a cross-data set facial expression recognition model and a recognition method, wherein, the reconstruction error of a training expression image is constrained by an adaptive non-negative weighting matrix, the function of important features in image data representation is strengthened, and the useless features with significant reconstruction error are reduced. In addition, by projecting the training set into the appropriate subspace through the mapping matrix, the intrinsic similarity between the cross-data set image samples can be better revealed, so that the subspace based on the low-rank and sparse representation can learn the robust reconstructed image in order to realize the final cross-data set facial expression recognition.

Description

technical field [0001] The invention relates to the field of facial expression recognition, in particular to a facial expression recognition model building and recognition method across data sets. Background technique [0002] Existing facial expression recognition methods are mainly based on the assumption that the captured test images and training images are from the same conditions and from the same individuals, and these methods are limited by a large number of labeled samples. However, when the training and testing processes of these methods come from expression databases of different conditions and individuals, their recognition performance will drop significantly. At the same time, when the labeled training samples are seriously insufficient, the existing facial expression recognition methods often appear overfitting in the learning process, so that the learned model cannot obtain the desired effect during the testing process. [0003] At present, only a few face rec...

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
CPCG06V40/174G06F18/23G06F18/2411G06F18/214
Inventor 马祥付俊妮
Owner CHANGAN 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