Affiliation:
1. Electronic Information Department, Dalian Polytechnic University, Dalian 116034, China
Abstract
Facial expression recognition using convolutional neural networks (CNNs) is a prevalent research area, and the network’s complexity poses obstacles for deployment on devices with limited computational resources, such as mobile devices. To address these challenges, researchers have developed lightweight networks with the aim of reducing model size and minimizing parameters without compromising accuracy. The LiteFer method introduced in this study incorporates depth-separable convolution and a lightweight attention mechanism, effectively reducing network parameters. Moreover, through comprehensive comparative experiments on the RAFDB and FERPlus datasets, its superior performance over various state-of-the-art lightweight expression-recognition methods is evident.
Funder
Special Funds for Basic Scientific Research Expenses of Undergraduate Universities in Liaoning Province
Reference29 articles.
1. Histograms of oriented gradients for human detection;Dalal;Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR’05),2005
2. Theory of communication;Gabor;J. Inst. Electr. Eng.,1946
3. A comparative study of texture measures with classification based on featured distributions;Ojala;Pattern Recognit.,1996
4. Pham, L., Vu, T.H., and Tran, T.A. (2021, January 10–15). Facial Expression Recognition Using Residual Masking Network. Proceedings of the 2020 25th International Conference on Pattern Recognition (ICPR), Milan, Italy.
5. Yao, A., Cai, D., Hu, P., Wang, S., Sha, L., and Chen, Y. (2016, January 12–16). Holonet: Towards robust emotion recognition in the wild. Proceedings of the 18th ACM International Conference on Multimodal Interaction, Tokyo, Japan.