Contrastive Mask Learning for Self-Supervised 3D Skeleton-Based Action Recognition
Sensors · 28.02.2025 · 25 (5) · 1521
DOI: 10.3390/s25051521 ↗Kurz gefasst
Fachartikel in der Zeitschrift Sensors, online erschienen am 28.02.2025. Verfasst von Haoyuan Zhang. Beteiligte Einrichtung: School of Electrical and Information Engineering, North Minzu Univeristy, Yinchuan 750021, China. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: High Level Introduction of Talent Research Start-up Fund of Ningxia Hui Autonomous Region, Fundamental Research Funds for Central Universities, North Minzu University, Ningxia Key Technologies and Scientific Innovation Team for Robotics and Intelligent Casting Equipment.
Abstract (Originalsprache)
In this paper, we propose a contrastive mask learning (CML) method for self-supervised 3D skeleton-based action recognition. Specifically, the mask modeling mechanism is integrated into multi-level contrastive learning with the aim of forming a mutually beneficial learning scheme from both contrastive learning and masked skeleton reconstruction. The contrastive objective is extended from an individual skeleton instance to clusters by closing the gap between cluster assignment from different instances of the same category, with the goal of pursuing inter-instance consistency. Compared with previous methods, CML integrates contrastive and masked learning comprehensively and enables intra-/inter-instance consistency pursuit via multi-level contrast, which leads to more discriminative skeleton representation learning. Our extensive evaluation of the challenging NTU RGB+D and PKU-MMD benchmarks demonstrates that representations learned via CML exhibit superior discriminability, consistently outperforming state-of-the-art methods in terms of action recognition accuracy.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- School of Electrical and Information Engineering, North Minzu Univeristy, Yinchuan 750021, China.
Förderung
- High Level Introduction of Talent Research Start-up Fund of Ningxia Hui Autonomous Region
- Fundamental Research Funds for Central Universities, North Minzu University
- Ningxia Key Technologies and Scientific Innovation Team for Robotics and Intelligent Casting Equipment
Zitieren
APA
Zhang, H. (2025). Contrastive Mask Learning for Self-Supervised 3D Skeleton-Based Action Recognition. Sensors, 25(5), 1521. https://doi.org/10.3390/s25051521
Vancouver
Zhang H. Contrastive Mask Learning for Self-Supervised 3D Skeleton-Based Action Recognition. Sensors. 2025;25(5):1521. doi:10.3390/s25051521
BibTeX
@article{Zhang2025,
author = {Zhang, Haoyuan},
title = {Contrastive Mask Learning for Self-Supervised 3D Skeleton-Based Action Recognition},
journal = {Sensors},
year = {2025},
volume = {25},
number = {5},
pages = {1521},
doi = {10.3390/s25051521}
}
EUVIMED