Fine-to-coarse self-attention graph convolutional network for skeleton-based action recognition

Basit Öğe Kaydı
dc.contributor.authorKilic, Ugur
dc.contributor.authorKaradag, Ozge Oztimur
dc.contributor.authorOzyer, Gulsah Tumuklu
dc.date.accessioned2026-01-24T12:31:07Z
dc.date.available2026-01-24T12:31:07Z
dc.date.issued2026
dc.departmentAlanya Alaaddin Keykubat Üniversitesi
dc.description.abstractSkeleton data has become an important modality in action recognition due to its robustness to environmental changes, computational efficiency, compact structure, and privacy-oriented nature. With the rise of deep learning, many methods for action recognition using skeleton data have been developed. Among these methods, spatial-temporal graph convolutional networks (ST-GCNs) have seen growing popularity due to the suitability of skeleton data for graph-based modeling. However, ST-GCN models use fixed graph topologies and fixed-size spatial-temporal convolution kernels. This limits their ability to model coordinated movements of joints in different body regions and long-term spatial-temporal dependencies. To address these limitations, we propose a fine-to-coarse self-attention graph convolutional network (FCSA-GCN). Our approach employs a fine-to-coarse scaling strategy for multi-scale feature extraction. This strategy effectively models both local and global spatial temporal relationships and better represents the interactions among joint groups in different body regions. By integrating a temporal self-attention mechanism (TSA) into the multi-scale feature extraction process, we enhance the model's ability to capture long-term temporal dependencies effectively. Additionally, during training, we employ the dynamic weight averaging (DWA) approach to ensure balanced optimization across the multi-scale feature extraction stages. Comprehensive experiments conducted on the NTU-60, NTU-120, and NW-UCLA datasets demonstrate that FCSA-GCN outperforms state-of-the-art methods. These results highlight that the proposed approach effectively addresses the current challenges in skeleton-based action recognition (SBAR).
dc.description.sponsorshipScientific and Technological Research Council of Turkey (TUBITAK) [124E309, 123E635]; Ataturk University [FDK-2023-11957]; TUBITAK
dc.description.sponsorshipThis study was supported by Scientific and Technological Research Council of Turkey (TUBITAK) under the Grant Numbers 124E309 and 123E635, and by Ataturk University under the BAP project code FDK-2023-11957. The authors thank to TUBITAK and University for their support.
dc.identifier.doi10.1016/j.asoc.2025.114268
dc.identifier.issn1568-4946
dc.identifier.issn1872-9681
dc.identifier.scopus2-s2.0-105022187562
dc.identifier.scopusqualityQ1
dc.identifier.urihttps://doi.org/10.1016/j.asoc.2025.114268
dc.identifier.urihttps://hdl.handle.net/20.500.12868/5645
dc.identifier.volume186
dc.identifier.wosWOS:001624435100009
dc.identifier.wosqualityQ1
dc.indekslendigikaynakWeb of Science
dc.indekslendigikaynakScopus
dc.language.isoen
dc.publisherElsevier
dc.relation.ispartofApplied Soft Computing
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanı
dc.rightsinfo:eu-repo/semantics/closedAccess
dc.snmzKA_WoS_20260121
dc.subjectFine-to-coarse approach
dc.subjectGraph convolutional networks
dc.subjectMulti-scale
dc.subjectSkeleton-based action recognition
dc.subjectSkeletal data
dc.subjectTemporal self-attention
dc.subjectmulti-scale feature extraction stages. Comprehensive experiments conducted on the NTU-60
dc.subjectNTU-120
dc.titleFine-to-coarse self-attention graph convolutional network for skeleton-based action recognition
dc.typeArticle

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu