Yanqiu Li, Shengzhao Li, Guangling Sun, Pu Yan. Lightweight Swin Transformer combined with multi-scale feature fusion for face expression recognition[J]. Opto-Electronic Engineering, 2025, 52(1): 240234
- Opto-Electronic Engineering
- Vol. 52, Issue 1, 240234 (2025)
Fig. 1. Swin Transformer network structure diagram
Fig. 2. Swin Transformer block module structure diagram
Fig. 3. Self-attention computing area. (a) MSA; (b) W-MSA; (c) SW-MSA
Fig. 4. Improved model structure diagram
Fig. 5. SPST module structure diagram
Fig. 6. A visual view of the BN, LN, and BCN standardization technology
Fig. 7. EMA module structure diagram
Fig. 8. Activation maps of the model before and after adding EMA module
Fig. 9. A partial sample of datasets
Fig. 10. Confusion matrix validation results on JAFFE. (a) Original Swin Transformer model; (b) Improved Swin Transformer model
Fig. 11. Confusion matrix validation results on RAF-DB. (a) Original Swin Transformer model; (b) Improved Swin Transformer model
Fig. 12. Confusion matrix validation results on FERPLUS. (a) Original Swin Transformer model; (b) Improved Swin Transformer model
Fig. 13. Confusion matrix validation results on FANE. (a) Original Swin Transformer model; (b) Improved Swin Transformer model
|
Table 1. Comparison of parameters before and after the model is improved
|
Table 2. Experimental comparison of replacing SPST modules in different stages
|
Table 3. Entropy comparison of activation maps
|
Table 4. Configuration of the experimental environment
| ||||||||||||||||||||||||||||||||||
Table 5. Accuracy of embedding the EMA module behind different stages
| |||||||||||||||||||||||||||||||||||||||||||
Table 6. Results of ablation experiments on FERPLUS, RAF-DB, and FANE
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Table 7. Accuracy comparsion of different networks on JAFFE，FERPLUS, and RAF-DB
Set citation alerts for the article
Please enter your email address
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20