A Review of Dynamic Facial Expression Recognition: Methods, Datasets and Directions

Xuantao Nie, Zixiang Fei, Wenju Zhou, Minrui Fei
Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing, PMLR 278:171-180, 2025.

Abstract

Dynamic facial expression recognition (DFER) has emerged as an essential area of research in computer vision, with implications in human-computer interaction, psychological analysis, and security. Although image-based static facial expression recognition (SFER) is well-developed, DFER captures temporal dynamics, remains less explored. This paper comprehensively reviews DFER, focusing on feature extraction methods from traditional handcrafted features to advanced deep learning techniques, analyzing performance metrics, and examining publicly available datasets with their comparative characteristics. We discuss specific challenges faced by DFER systems such as occlusion, pose variations, and temporal alignment. Finally, we explore promising applications in healthcare and human-computer interaction, providing concrete implementation strategies and future research directions.

Cite this Paper

BibTeX
@InProceedings{pmlr-v278-nie25a, title = {A Review of Dynamic Facial Expression Recognition: Methods, Datasets and Directions}, author = {Nie, Xuantao and Fei, Zixiang and Zhou, Wenju and Fei, Minrui}, booktitle = {Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing}, pages = {171--180}, year = {2025}, editor = {Zeng, Nianyin and Pachori, Ram Bilas and Wang, Dongshu}, volume = {278}, series = {Proceedings of Machine Learning Research}, month = {25--27 Apr}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v278/main/assets/nie25a/nie25a.pdf}, url = {https://proceedings.mlr.press/v278/nie25a.html}, abstract = {Dynamic facial expression recognition (DFER) has emerged as an essential area of research in computer vision, with implications in human-computer interaction, psychological analysis, and security. Although image-based static facial expression recognition (SFER) is well-developed, DFER captures temporal dynamics, remains less explored. This paper comprehensively reviews DFER, focusing on feature extraction methods from traditional handcrafted features to advanced deep learning techniques, analyzing performance metrics, and examining publicly available datasets with their comparative characteristics. We discuss specific challenges faced by DFER systems such as occlusion, pose variations, and temporal alignment. Finally, we explore promising applications in healthcare and human-computer interaction, providing concrete implementation strategies and future research directions.} }
Endnote
%0 Conference Paper %T A Review of Dynamic Facial Expression Recognition: Methods, Datasets and Directions %A Xuantao Nie %A Zixiang Fei %A Wenju Zhou %A Minrui Fei %B Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing %C Proceedings of Machine Learning Research %D 2025 %E Nianyin Zeng %E Ram Bilas Pachori %E Dongshu Wang %F pmlr-v278-nie25a %I PMLR %P 171--180 %U https://proceedings.mlr.press/v278/nie25a.html %V 278 %X Dynamic facial expression recognition (DFER) has emerged as an essential area of research in computer vision, with implications in human-computer interaction, psychological analysis, and security. Although image-based static facial expression recognition (SFER) is well-developed, DFER captures temporal dynamics, remains less explored. This paper comprehensively reviews DFER, focusing on feature extraction methods from traditional handcrafted features to advanced deep learning techniques, analyzing performance metrics, and examining publicly available datasets with their comparative characteristics. We discuss specific challenges faced by DFER systems such as occlusion, pose variations, and temporal alignment. Finally, we explore promising applications in healthcare and human-computer interaction, providing concrete implementation strategies and future research directions.
APA
Nie, X., Fei, Z., Zhou, W. & Fei, M.. (2025). A Review of Dynamic Facial Expression Recognition: Methods, Datasets and Directions. Proceedings of 2025 2nd International Conference on Machine Learning and Intelligent Computing, in Proceedings of Machine Learning Research 278:171-180 Available from https://proceedings.mlr.press/v278/nie25a.html.

Related Material