Lightweight facial expression recognition method based on deep convolution neural network

JunChen Tao; Shan Tian; Chao Chen

doi:10.1117/12.2648831

14 October 2022 Lightweight facial expression recognition method based on deep convolution neural network

JunChen Tao, Shan Tian, Chao Chen

Proceedings Volume 12343, 2nd International Conference on Laser, Optics and Optoelectronic Technology (LOPET 2022); 123432L (2022) https://doi.org/10.1117/12.2648831
Event: 2nd International Conference on Laser, Optics and Optoelectronic Technology (LOPET 2022), 2022, Qingdao, China

Abstract

On the basis of traditional algorithms, facial expression recognition combined with deep learning technology has improved the recognition accuracy and recognition speed in various scenes, especially in several large data sets. However, with the continuous vertical and horizontal expansion of the network structure focusing on the improvement of recognition effect, it has brought the problems of too large network volume, difficult deployment of mobile terminal, and inability to take into account the effect and practicability. Aiming at this problem, we proposed an improved vgg network structure, and processed the network structure, which greatly reduced the amount of network parameters under the condition of improving the recognition accuracy.

Citation Download Citation

JunChen Tao, Shan Tian, and Chao Chen "Lightweight facial expression recognition method based on deep convolution neural network", Proc. SPIE 12343, 2nd International Conference on Laser, Optics and Optoelectronic Technology (LOPET 2022), 123432L (14 October 2022); https://doi.org/10.1117/12.2648831

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available