Article Details

Title · Authors · Abstract · Keywords · DOI · References

ResNet-based Facial Expression Recognition Method Based on Adaptive Convolution and Attention Fusion

YAN Wujun, YE Jinxia, LI Jianchang

(College of Computer Science and Technology, Taiyuan Normal University, Jinzhong 030619, China)

Section：Artificial Intelligence · Vol. 2025, Issue 5 · Pages：39-44,50 · 2096-4706（2025）05-0039-07 · CLC number：TP183 · Document code：A · DOI：10.19850/j.cnki.2096-4706.2025.05.007

Abstract

In practical applications, the accuracy of facial expression recognition is relatively low because facial images are affected by factors such as lighting, occlusion, and pose. To address this, an expression recognition method that integrates adaptive convolution and the Attention Mechanism is proposed. Based on the ResNet34 network, this method introduces an AKConv module to capture multi-scale features and integrates an ACmix mechanism to improve the classification accuracy. Meanwhile, the traditional cross-entropy loss function is replaced with SlideLoss to address the problem of data imbalance. Experimental results show that this model achieves an accuracy of 75.21% on the FER2013 dataset, verifying its effectiveness and superiority and providing new ideas and methods for the field of facial expression recognition.

Keywords

ResNet; facial emotion recognition; Deep Learning; Attention Mechanism

References

  1. [1] Zhao Xiao, Yang Chen, Wang Ruonan, etc Facial expression recognition based on attention mechanism ResNet lightweight network [J]. Liquid Crystal and Display, 2023, 38 (11): 1503-1510
  2. [2] MEHRABIAN A，WEINER P，GOODENOUGH L. Explorations in Personality [M].New York：Wiley，1971：84-96.
  3. [3] EKMAN P，FRIESEN W V. Constants Across Cultures in the Face and Emotion [J].Journal of Personality and Social Psychology，1971，17（2）：124-129.
  4. [4] DALAL N，TRIGGS B. Histograms of Oriented Gradients for Human Detection [C]//Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).Washington：IEEE，2005：886-893.
  5. [5] OJALA T，PIETIKAINEN M. Multiresolution Gray-Scale and Rotation Invariant Texture Classification with Local Binary Patterns [J].IEEE Transactions on Pattern Analysis and Machine Intelligence，2002，24（7）：971-987.
  6. [6] TURK M，PENTLAND A. Eigenfaces for Recognition [J].Journal of Cognitive Neuroscience，1991，3（1）：71-86.
  7. [7] LECUN Y，BENGIO Y，HINTON G. Deep Learning [J].Nature，2015，521（7553）：436-444.
  8. [8] HINTON G，SUTSKEVER I，TARVALSEN A. Improving Neural Networks by Preventing Co-Adaptation of Feature Detectors [J/OL].arXiv:1207.0580 [cs.NE].（2012-07-03）.https://arxiv.org/abs/1207.0580?context=cs.LG.
  9. [9] KRIZHEVSKY A，SUTSKEVER I，HINTON G. ImageNet Classification with Deep Convolutional Neural Networks [C]//Proceedings of the 25th International Conference on Neural Information Processing Systems-Volume 1 (NIPS'12).Lake Tahoe：Curran Associates Inc，2012：1097-1105.
  10. [10] SIMONYAN K，ZISSERMAN A. Very Deep Convolutional Networks for Large-Scale Image Recognition [J/OL].arXiv:1409.1556 [cs.CV]（2015-04-10）. https://arxiv.org/abs/1409.1556.
  11. [11] Yuan Lang Detection and Vectorization of Land Object Changes Based on Deep Learning [D]. Chengdu: University of Electronic Science and Technology of China, 2021
  12. [12] Guo Xingang, Cheng Chao, Shen Ziqi Facial expression recognition based on convolutional network attention mechanism [J]. Journal of Jilin University: Engineering Edition, 2024, 54 (8): 2319-2328
  13. [13] Shi Cuiping, Tan Cong, Zuo Jiang, etc Facial expression recognition based on improved AlexNet convolutional neural network [J]. Telecommunications Technology, 2020, 60 (9): 1005-1012
  14. [14] Peng Qi, Liu Yinhua, Shang Yunrui Facial expression recognition method integrating key region attention mechanism [J/OL]. Computer Engineering, 2024:1-10 [2022-08-21] https://doi.org/10.19678/j.issn.1000-3428.00EC0069661.
  15. [15] Zhang Bocheng, Li Wei Facial expression recognition based on feature fusion and attention mechanism [J]. Changjiang Information and Communication, 2024, 37 (4): 17-20+35

About the authors

Yan Wujun (1973-), male, Han ethnicity, from Houma, Shanxi, associate professor, master's supervisor, master's degree, research direction: artificial intelligence, deep learning, machine learning; Ye Jinxia (1999-), female, Han ethnicity, Shanxi University Tongren, currently pursuing a master's degree in deep learning and image processing; Li Jianchang (1998-), male, Han ethnicity, from Dongying, Shandong, is currently pursuing a master's degree in deep learning, medical image processing, and computer vision.