Research on facial expression recognition algorithm based on convolutional neural network

Xiaobo Zhang, Yuliang Yang*, Linhao Zhang, Wanchong Li, Shuai Dang, Peng Wang, Mengyu Zhu

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

2 Citations (Scopus)

Abstract

A network model for facial expression recognition is designed and named DI-FERNet in this paper. The network uses depth-wise separable convolution, dilated convolution and residual module to build the network structure. This paper uses MTCNN to perform face alignment processing on the pictures in the dataset. A large number of experiments are carried out on the selected expression datasets KDEF and RAF. The test accuracy on KDEF is 97.2% and on the RAF is 77.1%.

Original languageEnglish
Title of host publication2019 28th Wireless and Optical Communications Conference, WOCC 2019 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728106601
DOIs
Publication statusPublished - May 2019
Event28th Wireless and Optical Communications Conference, WOCC 2019 - Beijing, China
Duration: 9 May 201910 May 2019

Publication series

Name2019 28th Wireless and Optical Communications Conference, WOCC 2019 - Proceedings

Conference

Conference28th Wireless and Optical Communications Conference, WOCC 2019
Country/TerritoryChina
CityBeijing
Period9/05/1910/05/19

Keywords

  • Depth-wise separable convolution
  • Dilated convolution
  • FER
  • MTCNN

Fingerprint

Dive into the research topics of 'Research on facial expression recognition algorithm based on convolutional neural network'. Together they form a unique fingerprint.

Cite this

Zhang, X., Yang, Y., Zhang, L., Li, W., Dang, S., Wang, P., & Zhu, M. (2019). Research on facial expression recognition algorithm based on convolutional neural network. In 2019 28th Wireless and Optical Communications Conference, WOCC 2019 - Proceedings Article 8770616 (2019 28th Wireless and Optical Communications Conference, WOCC 2019 - Proceedings). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/WOCC.2019.8770616