EUVIMEDEUVIMED entdecken
Crossref

An Enhanced Adaptive Confidence Margin for Semi-Supervised Facial Expression Recognition

Hangyu Li, Nannan Wang, Xi Yang, Xiaoyu Wang, Xinbo Gao

IEEE Transactions on Pattern Analysis and Machine Intelligence · 2026-02 · 48 (2) · 1278-1290

DOI: 10.1109/tpami.2025.3612953 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2026-02. Verfasst von Hangyu Li und 4 weiteren Autor:innen. Hauptthemen laut MEDLINE: Facial Expression, Supervised Machine Learning, Image Processing, Computer-Assisted, Automated Facial Recognition. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 74 Literaturangaben. Gefördert von: National Natural Science Foundation of China, Scientific and Technological Innovation Teams in Shaanxi Province, Shaanxi Province Core Technology Research and Development Project.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Semi-supervised learning (SSL) provides a practical framework for leveraging massive unlabeled samples, especially when labels are expensive for facial expression recognition (FER). Typical SSL methods like FixMatch select unlabeled samples with confidence scores above a fixed threshold for training. However, these methods face two primary limitations: failing to consider the varying confidence across facial expression categories and failing to utilize unlabeled facial expression samples efficiently. To address these challenges, we propose an Enhanced Adaptive Confidence Margin (EACM), consisting of dynamic thresholds for different categories, to fully learn unlabeled samples. Specifically, we employ the predictions on labeled samples at each training iteration to learn an EACM. It then partitions unlabeled samples into two subsets: (1) subset I, including samples whose confidence scores are no less than the margin; (2) subset II, including samples whose confidence scores are less than the margin. For samples in subset I, we constrain their predictions on strongly-augmented versions to match the pseudo-labels derived from the predictions on weakly-augmented versions. Meanwhile, we introduce a feature-level contrastive objective to enhance the similarity between two weakly-augmented features of a sample in subset II. We extensively evaluate EACM on image-based and video-based facial expression datasets, showing that our method achieves superior performance, significantly surpassing fully-supervised baselines in a semi-supervised manner. Additionally, our EACM is promising to leverage cross-dataset unlabeled samples for practical training to boost fully-supervised performance.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Facial ExpressionSupervised Machine LearningImage Processing, Computer-AssistedAutomated Facial RecognitionHumansFaceAlgorithmsDatabases, Factual

Förderung

  • National Natural Science Foundation of China
  • Scientific and Technological Innovation Teams in Shaanxi Province
  • Shaanxi Province Core Technology Research and Development Project
  • Fundamental Research Funds for the Central Universities

Zitieren

APA

Li, H., Wang, N., Yang, X., Wang, X., & Gao, X. (2026). An Enhanced Adaptive Confidence Margin for Semi-Supervised Facial Expression Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 48(2), 1278-1290. https://doi.org/10.1109/tpami.2025.3612953

Vancouver

Li H, Wang N, Yang X, Wang X, Gao X. An Enhanced Adaptive Confidence Margin for Semi-Supervised Facial Expression Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2026;48(2):1278-1290. doi:10.1109/tpami.2025.3612953

BibTeX
@article{Li2026,
  author = {Li, Hangyu and Wang, Nannan and Yang, Xi and Wang, Xiaoyu and Gao, Xinbo},
  title = {An Enhanced Adaptive Confidence Margin for Semi-Supervised Facial Expression Recognition},
  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year = {2026},
  volume = {48},
  number = {2},
  pages = {1278--1290},
  doi = {10.1109/tpami.2025.3612953}
}
Kostenlos starten →