Facial Expression Recognition Method Based on Residual Masking Reconstruction Network
2
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Facial expression recognition can aid in the development of fatigue driving detection, teaching qualityevaluation, and other fields. In this study, a facial expression recognition method was proposed with a residualmasking reconstruction network as its backbone to achieve more efficient expression recognition andclassification. The residual layer was used to acquire and capture the information features of the input image,and the masking layer was used for the weight coefficients corresponding to different information features toachieve accurate and effective image analysis for images of different sizes. To further improve the performanceof expression analysis, the loss function of the model is optimized from two aspects, feature dimension anddata dimension, to enhance the accurate mapping relationship between facial features and emotional labels. Thesimulation results show that the ROC of the proposed method was maintained above 0.9995, which canaccurately distinguish different expressions. The precision was 75.98%, indicating excellent performance ofthe facial expression recognition model.
Abstract
Facial expression recognition can aid in the development of fatigue driving detection, teaching qualityevaluation, and other fields. In this study, a facial expression recognition method was proposed with a residualmasking reconstruction network as its backbone to achieve more efficient expression recognition andclassification. The residual layer was used to acquire and capture the information features of the input image,and the masking layer was used for the weight coefficients corresponding to different information features toachieve accurate and effective image analysis for images of different sizes. To further improve the performanceof expression analysis, the loss function of the model is optimized from two aspects, feature dimension anddata dimension, to enhance the accurate mapping relationship between facial features and emotional labels. Thesimulation results show that the ROC of the proposed method was maintained above 0.9995, which canaccurately distinguish different expressions. The precision was 75.98%, indicating excellent performance ofthe facial expression recognition model.
참고문헌 (10)
- A. R. Hazourli, "Multi-facial patches aggregation network for facial expression recognition and facial regions contributions to emotion display", Multimedia Tools and Applications, 80, pp.13639~13662 (2021)
- B, Yang, "Facial expression recognition based on multi-dataset neural network", Radioengineering, 29(1), pp.259~266 (2020)
- X. Zhang, "Joint expression synthesis and representation learning for facial expression recognition", IEEE Transactions on Circuits and Systems for Video Technology, 32(3), pp.1681~1695 (2022)
- C. V. R. Reddy, "Facial emotion recognition using NLPCA and SVM", Traitement du Signal, 36(1), pp.13~22 (2019)
- D. Poux, "Dynamic facial expression recognition under partial occlusion with optical flow reconstruction", IEEE Transactions on Image Processing, 31, pp.446~457 (2022)
- C. Xu, "Person-independent facial expression recognition method based on improved Wasserstein generative adversarial networks in combination with identity aware", Multimedia Systems, 26, pp.53~61 (2020)
- K. Talele, "Facial expression recognition using digital signature feature descriptor", Signal, Image and Video Processing, 14, pp.701~709 (2020)
- Y. Yaermaimaiti, "Statement of Removal", Journal of Decision Systems, pp.1~1 (2021), Informa UK Limited[DOI]
- I. Gogic, "Fast facial expression recognition using local binary features and shallow neural networks", The Visual Computer, 36, pp.97~112 (2020)
- L. Pham, "Facial expression recognition using residual masking network", Proceedings of 2020 25th International Conference on Pattern Recognition (ICPR), pp.4513~4519 (2021)