Technology for Facial Expression Recognition Across Different Scenes
Main Article Content
Keywords
facial expression recognition, environmental robustness, lightweight, domain adaptation, cross-scene
Abstract
This paper conducts a systematic review and analysis of the key technologies for facial expression recognition across different scenarios. As the facial expression recognition technology moves from the laboratory to diverse real -world scenarios such as remote healthcare, mental health , and intelligent cabins, ensuring the model maintains high performance in complex environments has become a core issue that needs to be urgently addressed. Firstly, this paper reviews three key technology systems focusing on environmental robustness, efficient deployment, and data scarcity and domain adaptation. It elaborates on representative methods such as attention mechanisms, lightweight networks, and domain adaptation, along with their progress. Based on this, it further analyzes the common challenge s faced in cross -scenario applications, including environmental interference, data distribution differences, and deployment constraints. The review results indicate that there is no universal optimal model. The technical solutions need to be deeply adapted to specific scene requirements. The integration of robustness, lightweighting, and domain adaptation technologies is the future development trend. This paper provides theoretical basis and practical references for the selection of facial expression recognition technologies for specific scenarios.
References
- [1] Liu, H., Cai, H., Li, Q., et al. (2022). Adaptive Multilayer Perceptual Attention Network for Facial Expression Recognition. IEEE Transactions on Circuits and Systems for Video Technology, 32 (9), 6253-6266.
- [2] Liu, C., Liu, X., Chen, C., et al. (2024). Deep Global Multiple -Scale and Local Patches Attention D ual- Branch Network for Pose -Invariant Facial Expression Recognition. Computer Modeling in Engineering & Sciences, 139(1), 405-440.
- [3] Huang, Y., Peng, J., Zhang, W., et al. (2025). FERMixNet: An Occlusion Robust Facial Expression Recognition Model With Facial Mixing Augmentation and Mid -Level Representation Learning. IEEE Transactions on Affective Computing, 16(2), 639-654.
- [4] Ma, H., Lei, S., Li, H. C., et al. (2025). FER -VMamba: A robust facial expression recognition framework with global compact attention and hierarchical feature interaction. Information Fusion, 124, 103371.
- [5] Zhao, Z., Li, Y., Yang, J., et al. (2024). A lightweight facial expression recognition model for automated engagement detection. Signal, Image and Video Processing, 18(4), 3553-3563.
- [6] Mou, X., Song, Y., Wang, R., et al. (2023). Lightweight Facial Expression Recognition Based on Class - Rebalancing Fusion Cumulative Learning. Applied Sciences, 13(15), 9029.
- [7] Yang, J., Qian, T., Zhang, F., et al. (2021). Real -Time Facial Expres sion Recognition Based on Edge Computing. IEEE Access, 9, 76178-76190.
- [8] Riaz, W., Ji, J., & Ullah, A. (2025). TriViT -Lite: A Compact Vision Transformer –MobileNet Model with Texture -Aware Attention for Real -Time Facial Emotion Recognition in Healthcare. Electronics, 14(16), 3256.
- [9] Li, S., & Deng, W. (2022). A Deeper Look at Facial Expression Dataset Bias. IEEE Transactions on Affective Computing, 13(2), 881-893.
- [10] Yan, J., Du, C., Li, B., et al. (2025). Cross-database facial expression recognition based on Multi-feature Representation Multi -layer Domain Adaptive Fusion Network. Engineering Applications of Artificial Intelligence, 155, 110995.
- [11] Kim, H., Jung, Y., & Song, B. C. (2025). DIRE: Enhancing Facial Expression Recognition Through Domain-Invariant Representation Learning for Robust Generalization. IEEE Transactions on Multimedia, 27, 9542-9554.
- [12] Chen, T., Pu, T., Wu, H., et al. (2022). Cross -Domain Facial Expression Recognition: A Unified Evaluation Benchmark and Adversarial Graph Learni ng. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(12), 9887-9903.
Article Sidebar
This work is licensed under a Creative Commons Attribution 4.0 International License.