Facial Expression Recognition Method based on Squeeze and Excitation Module

Yingya Zheng ； Shiwei Xu ； Xinglei Xu

《International Core Journal of Engineering》 7卷3期 (2021/03) Pp. 433-439

https://doi.org/10.6919/ICJE.202103_7(3).0059

摘要

In order to improve the accuracy of facial expression recognition, this paper proposes a convolutional neural network expression recognition method based on squeeze and excitation module. This method first serially merges the multi-layer features of the VGG network to extract more comprehensive features; Then integrate the SE module into the improved network to make the network automatically learn the importance of each feature channel, and then improve the useful features according to this importance to improve the classification accuracy; Finally, the fully connected layer is designed, and the 7-dimensional vector is directly output and classified using softmax. The experimental results show that the method proposed in this paper achieves a higher recognition rate.

關鍵字

Deep Learning ； Expression Recognition ； Convolutional Network ； Visual Geometry Group ； Squeeze and Excitation Network

參考文獻

A. Mollahosseini, D. Chan and M. H. Mahoor, "Going deeper in facial expression recognition using deep neural networks," 2016 IEEE Winter Conference on Applications of Computer Vision (WACV), Lake Placid, NY, 2016, pp. 1-10, doi: 10.1109/WACV.2016.7477450.

Kim B K, Roh J, Dong S Y, et al. Hierarchical committee of deep convolutional neural networks for robust facial expression recognition[J]. Journal on Multimodal User Interfaces, 2016, 10(2):1-17.

Goodfellow I.J. et al. (2013) Challenges in Representation Learning: A Report on Three Machine Learning Contests. In: Lee M., Hirose A., Hou ZG., Kil R.M. (eds) Neural Information Processing. ICONIP 2013. Lecture Notes in Computer Science, vol 8228. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-42051-1_16

Lucey P, Cohn J F, Kanade T, et al. The Extended Cohn-Kanade Dataset (CK+): A complete dataset for action unit and emotion-specified expression[C]// Computer Vision & Pattern Recognition Workshops. IEEE, 2010.

Russakovsky O, Deng J, et al. Imagenet large scale visual recognition challenge[C]// International Journal of Computer Vision, 2015,115(3): 211-252.