QCS:Feature Refining from Quadruplet Cross Similarity for Facial Expression Recognition
Proceedings of the AAAI Conference on Artificial Intelligence · 11.04.2025 · 39 (7) · 7563-7572
DOI: 10.1609/aaai.v39i7.32814 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 11.04.2025. Verfasst von Chengpeng Wang und 4 weiteren Autor:innen. Die Arbeit wurde bisher 17-mal zitiert (Crossref).
Abstract (Originalsprache)
Facial expression recognition faces challenges where labeled significant features in datasets are mixed with unlabeled redundant ones. In this paper, we introduce Cross Similarity Attention (CSA) to mine richer intrinsic information from image pairs, overcoming a limitation when the Scaled Dot-Product Attention of ViT is directly applied to calculate the similarity between two different images. Based on CSA, we simultaneously minimize intra-class differences and maximize inter-class differences at the fine-grained feature level through interactions among multiple branches. Contrastive residual distillation is utilized to transfer the information learned in the cross module back to the base network. We ingeniously design a four-branch centrally symmetric network, named Quadruplet Cross Similarity (QCS), which alleviates gradient conflicts arising from the cross module and achieves balanced and stable training. It can adaptively extract discriminative features while isolating redundant ones. The cross-attention modules exist during training, and only one base branch is retained during inference, resulting in no increase in inference time. Extensive experiments show that our proposed method achieves state-of-the-art performance on several FER datasets.
Zitieren
APA
Wang, C., Chen, L., Wang, L., Li, Z., & Lv, X. (2025). QCS:Feature Refining from Quadruplet Cross Similarity for Facial Expression Recognition. Proceedings of the AAAI Conference on Artificial Intelligence, 39(7), 7563-7572. https://doi.org/10.1609/aaai.v39i7.32814
Vancouver
Wang C, Chen L, Wang L, Li Z, Lv X. QCS:Feature Refining from Quadruplet Cross Similarity for Facial Expression Recognition. Proceedings of the AAAI Conference on Artificial Intelligence. 2025;39(7):7563-7572. doi:10.1609/aaai.v39i7.32814
BibTeX
@article{Wang2025,
author = {Wang, Chengpeng and Chen, Li and Wang, Lili and Li, Zhaofan and Lv, Xuebin},
title = {QCS:Feature Refining from Quadruplet Cross Similarity for Facial Expression Recognition},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2025},
volume = {39},
number = {7},
pages = {7563--7572},
doi = {10.1609/aaai.v39i7.32814}
}
EUVIMED