EUVIMEDEUVIMED entdecken
Crossref

ReSup: Reliable Label Noise Suppression for Facial Expression Recognition

Xiang Zhang, Yan Lu, Huan Yan, Jinyang Huang, Yu Gu, Yusheng Ji, Zhi Liu, Bin Liu

IEEE Transactions on Affective Computing · 2025-07 · 16 (3) · 2006-2019

DOI: 10.1109/taffc.2025.3549017 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Affective Computing, 2025-07. Verfasst von Xiang Zhang und 7 weiteren Autor:innen. Die Arbeit wurde bisher 19-mal zitiert (Crossref) und nennt 70 Literaturangaben. Gefördert von: Anhui Province Science Foundation for Youths, Major scientific and technological project of Anhui Provincial Science and Technology Innovation Platform, National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Because of the ambiguous and subjective property of the facial expression, the label noise is widely existing in the FER dataset. For this problem, in the training phase, current methods often directly predict whether the label is noised or not, aiming to reduce the contribution of the noised data. However, we argue that this kind of method suffers from the low reliability of such noise data decision operation. It makes that some mistakenly abounded clean data are not utilized sufficiently and some mistakenly kept noised data disturbing the model learning. In this paper, we propose a more reliable noise-label suppression method called ReSup. First, instead of directly predicting noised or not, ReSup makes the noise data decision by modeling the distribution of noise and clean labels simultaneously according to the disagreement between the prediction and the target. Specifically, to achieve optimal distribution modeling, ReSup models the similarity distribution of all samples. To further enhance the reliability of our noise decision results, ReSup uses two networks to jointly achieve noise suppression. Specifically, ReSup utilize the property that two networks are less likely to make the same mistakes, making two networks swap decisions and tending to trust decisions with high agreement. Extensive experiments on popular datasets shows the effectiveness of ReSup.

Förderung

  • Anhui Province Science Foundation for Youths
  • Major scientific and technological project of Anhui Provincial Science and Technology Innovation Platform
  • National Natural Science Foundation of China
  • Guizhou Province Basic Research Plan Youth Guidance Project, China

Zitieren

APA

Zhang, X., Lu, Y., Yan, H., Huang, J., Gu, Y., Ji, Y., Liu, Z., & Liu, B. (2025). ReSup: Reliable Label Noise Suppression for Facial Expression Recognition. IEEE Transactions on Affective Computing, 16(3), 2006-2019. https://doi.org/10.1109/taffc.2025.3549017

Vancouver

Zhang X, Lu Y, Yan H, Huang J, Gu Y, Ji Y, et al. ReSup: Reliable Label Noise Suppression for Facial Expression Recognition. IEEE Transactions on Affective Computing. 2025;16(3):2006-2019. doi:10.1109/taffc.2025.3549017

BibTeX
@article{Zhang2025,
  author = {Zhang, Xiang and Lu, Yan and Yan, Huan and Huang, Jinyang and Gu, Yu and Ji, Yusheng and Liu, Zhi and Liu, Bin},
  title = {ReSup: Reliable Label Noise Suppression for Facial Expression Recognition},
  journal = {IEEE Transactions on Affective Computing},
  year = {2025},
  volume = {16},
  number = {3},
  pages = {2006--2019},
  doi = {10.1109/taffc.2025.3549017}
}
Kostenlos starten →