EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Semantics-Assisted Training Graph Convolution Network for Skeleton-Based Action Recognition

Huangshui Hu, Yu Cao, Yue Fang, Zhiqiang Meng

Sensors · 15.03.2025 · 25 (6) · 1841

DOI: 10.3390/s25061841 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 15.03.2025. Verfasst von Huangshui Hu und 3 weiteren Autor:innen. Beteiligte Einrichtung: College of Computer Science and Engineering, Changchun University of Technology, Changchun 130012, China. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 49 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the science and technology development project of Jilin province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The skeleton-based action recognition networks often focus on extracting features such as joints from samples, while neglecting the semantic relationships inherent in actions, which also contain valuable information. To address the lack of utilization of semantic information, this paper proposes a semantics-assisted training graph convolution network (SAT-GCN). By dividing the features outputted by the skeleton encoder into four parts and contrasting them with the text features generated by the text encoder, the obtained contrastive loss is used to guide the overall network training. This approach effectively improves recognition accuracy while reducing the number of model parameters. In addition, angle features are incorporated into the skeleton model input to aid in classifying similar actions. Finally, a multi-feature skeleton encoder is designed to separately extract features such as joints, bones, and angles. These extracted features are then integrated through feature fusion. The fused features are then passed through three graph convolution blocks before being fed into fully connected layers for classification. Extensive experiments were conducted on three large-scale datasets, NTU RGB + D 60, NTU RGB + D 120, and NW-UCLA to validate the performance of the proposed model. The results show that the SAT-GCN outperforms others in terms of both accuracy and number of parameters.

Schlagwörter

Stichwörter der Autor:innen

action recognitionfeature fusionsemantic relationshipssemantics-assisted training

Autor:innen und Institutionen

  • College of Computer Science and Engineering, Changchun University of Technology, Changchun 130012, China.

Förderung

  • the science and technology development project of Jilin province

Zitieren

APA

Hu, H., Cao, Y., Fang, Y., & Meng, Z. (2025). Semantics-Assisted Training Graph Convolution Network for Skeleton-Based Action Recognition. Sensors, 25(6), 1841. https://doi.org/10.3390/s25061841

Vancouver

Hu H, Cao Y, Fang Y, Meng Z. Semantics-Assisted Training Graph Convolution Network for Skeleton-Based Action Recognition. Sensors. 2025;25(6):1841. doi:10.3390/s25061841

BibTeX
@article{Hu2025,
  author = {Hu, Huangshui and Cao, Yu and Fang, Yue and Meng, Zhiqiang},
  title = {Semantics-Assisted Training Graph Convolution Network for Skeleton-Based Action Recognition},
  journal = {Sensors},
  year = {2025},
  volume = {25},
  number = {6},
  pages = {1841},
  doi = {10.3390/s25061841}
}
Kostenlos starten →