Recognition of Teachers’ Facial Expression Intensity Based on Convolutional Neural Network and Attention Mechanism

Dong Yang,Jinling Cui,Kun Zheng,Junhua Liu

doi:10.1109/access.2020.3046225

Abstract

The evaluations of traditional teaching quality are mainly subjective, and there is a lack of fine-grained objective data to support the evaluation of teaching states in the classroom. In this paper, an intensity-based facial expression dataset is proposed and named EIDB-13, which contains 13 kinds and 10393 facial images collected from thousands of individuals and existing facial expression datasets. Convolutional neural network (CNN) and attention mechanism are combined to recognize facial expressions. Migration learning is used to solve over-fitting problem in the process of training deep network based on the small sample dataset. InceptionResNetV2 is employed as migration network. Furthermore, an InceptionResNetV2+CBAM network proposed extract similar feature information among facial expressions and it outperforms the network without attention mechanisms. Experiments show a classification accuracy rate of 78% on the intensity-based facial expression dataset EIDB-13 and of 88% on the public macro expression dataset RAF-DB. Combining facial expression recognition technology into teaching is a key foundation to study teaching quality on the intensity of teacher’s expression.

Highlights

In the traditional classroom, students’ knowledge acquisitions are inseparable from the teacher’s performance
Compared with other expression recognition methods, the result of training on RAF-DB dataset with CBAM attention mechanism added to the InceptionResNetV2 network is better
A recognition algorithm based on convolutional neural network InceptionResNetV2 and attention mechanism module CBAM is proposed

Summary

Introduction

Students’ knowledge acquisitions are inseparable from the teacher’s performance. Relevant practitioners in the field of education began to reach a consensus that teacher evaluation is the key factor to improve the quality of teaching and professional development [1]. Colleges mainly use the method of anonymous evaluation to assess teachers’ performance. The age, gender, skin color and even attractiveness of teachers have an impact on the evaluation of teaching, which has strong subjectivity and blindness [2], [3]. Due to the epidemic COVID-19, E-Learning has been more widely practiced than ever before. The teaching evaluation method has not been applied to E-Learning. Regardless of traditional or online teaching mode, the teaching evaluation methods are mainly subjective, and lack the support of finegrained objective data

Results

Discussion

Conclusion