Facial Expression Recognition Through Joint Modeling of Local Visual Representations and Continuous Affect
Sensors · 18.11.2026 · 26 (22) · 7279
DOI: 10.3390/s26227279 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 18.11.2026. Verfasst von Pujun Xue und 5 weiteren Autor:innen. Die Arbeit nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Jiangsu Provincial University Key Laboratory of Intelligent Multi-source Information Processing and Security.
Abstract (Originalsprache)
Facial expressions are an important means of conveying human emotions. Accurately distinguishing similar expressions requires capturing subtle appearance differences across facial regions, while a single class label cannot fully represent the continuous affective relationships among expression categories. To this end, we propose a Valence-Arousal-Guided Region Attention Network (VA-RANet) for joint discrete expression classification and continuous valence-arousal (VA) estimation. The model learns multiple spatial weight maps from shared features, constructs regional representations through weighted aggregation, and fuses them with global features. A region complementarity constraint is employed to limit the similarity among the spatial weight maps and encourage differentiated regional responses. During training, the distances between the samples and the fixed class prototypes in the VA space are converted into a probability distribution and subsequently fused with the discrete expression labels to generate soft labels that provide auxiliary supervision for discrete expression classification. Ground-truth VA annotations are used for AffectNet, whereas VA pseudo-labels obtained through transfer are used for RAF-DB and CAER-S. Only facial images are required at inference. The model achieves classification accuracies of 65.71% and 62.93% under the seven-class and eight-class AffectNet settings, respectively. Experimental results indicate that the combination of differentiated multi-region visual representations and VA-guided supervision contributes to improved facial expression recognition performance.
Förderung
- Jiangsu Provincial University Key Laboratory of Intelligent Multi-source Information Processing and Security
Zitieren
APA
Xue, P., Sun, Q., Xu, W., Lin, M., Cangelosi, A., & Cai, H. (2026). Facial Expression Recognition Through Joint Modeling of Local Visual Representations and Continuous Affect. Sensors, 26(22), 7279. https://doi.org/10.3390/s26227279
Vancouver
Xue P, Sun Q, Xu W, Lin M, Cangelosi A, Cai H. Facial Expression Recognition Through Joint Modeling of Local Visual Representations and Continuous Affect. Sensors. 2026;26(22):7279. doi:10.3390/s26227279
BibTeX
@article{Xue2026,
author = {Xue, Pujun and Sun, Qigao and Xu, Wei and Lin, Mufang and Cangelosi, Angelo and Cai, Huili},
title = {Facial Expression Recognition Through Joint Modeling of Local Visual Representations and Continuous Affect},
journal = {Sensors},
year = {2026},
volume = {26},
number = {22},
pages = {7279},
doi = {10.3390/s26227279}
}
EUVIMED