Client-Unbiased Skeletal Action Recognizer in Federated Learning
IEEE Transactions on Image Processing · 2025 · 34 · 4908-4922
DOI: 10.1109/tip.2025.3586511 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Image Processing, 2025. Verfasst von Xingyu Zhu und 2 weiteren Autor:innen. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 54 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Edge sensor devices generate vast amounts of user data, but centralized processing poses privacy risks. Federated Learning addresses this by decentralizing training. However, applying Federated Learning directly to skeleton videos fails to preserve motion dynamics and suffers from client heterogeneity bias. To address these limitations, we propose CSAR-a Client-Unbiased Skeletal Action Recognizer for Federated Learning-which tackles two core challenges: motion dynamics preservation and classifier bias mitigation. Specifically, CSAR employs a Model Calibration Loss during client training to align client-server representations and reduce drift. On the server, it generates class-balanced spatiotemporal federated features through Prototypical Gaussian Sampling, subsequently refined via a Motion-aware Differential Loss to capture kinematic properties. These features enable retraining of a globally debiased recognizer that achieves accuracy comparable to real-data-trained models. Further stabilization is achieved through Knowledge Matching, which enhances global understanding. Experiments under natural and label heterogeneity confirm that CSAR outperforms state-of-the-art methods.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Zhu, X., Shu, X., & Tang, J. (2025). Client-Unbiased Skeletal Action Recognizer in Federated Learning. IEEE Transactions on Image Processing, 34, 4908-4922. https://doi.org/10.1109/tip.2025.3586511
Vancouver
Zhu X, Shu X, Tang J. Client-Unbiased Skeletal Action Recognizer in Federated Learning. IEEE Transactions on Image Processing. 2025;34:4908-4922. doi:10.1109/tip.2025.3586511
BibTeX
@article{Zhu2025,
author = {Zhu, Xingyu and Shu, Xiangbo and Tang, Jinhui},
title = {Client-Unbiased Skeletal Action Recognizer in Federated Learning},
journal = {IEEE Transactions on Image Processing},
year = {2025},
volume = {34},
pages = {4908--4922},
doi = {10.1109/tip.2025.3586511}
}
EUVIMED