Aug 2026· International Conference Computational Vision and Bio Inspired Computing· pp. 635-639· 0 citations· 17 references
Abstract
There has been a lot of development in “skeleton-based human activity recognition” due to its emphasis on human joint actions, which eliminates the effect of background noises such as lighting, body appearance changes, or other environmental aspects. However, there are limitations in the skeleton data that can be caused by various factors, including occlusions, missing joints, appearance loss, skeleton overlap, and inaccurate estimates of human actions. In general, existing Graph Convolutional Network (GCN) and Transformer approaches perform poorly with incomplete skeletons. To solve the problem, this paper introduces “Occlusion-Aware GCN-Transformer Framework for robust skeleton-based human activity recognition”. The framework uses an adaptive spatial GCN to detect local joint relationships and a spatiotemporal transformer to identify motion dependencies between frames. To augment model robustness, the training and evaluation processes encompass various occlusion scenarios, including random joint masking, upper-body occlusion, lower-body occlusion, and corrupted temporal frames.
Skeleton-based human action recognition has attracted increasing attention because it provides a concise structural representation of human motion and is less affected by environmental factors. However, for real-deployment scenarios, the skeleton data are usually extracted automatically using human pose estimation mode...
Khac-Anh Phu, Van-Dung Hoang, Van-Tuong-Lan Le· HUE UNIVERSITY JOURNAL OF SC...· 0 citations
A novel approach called Spatio-Temporal Local Graph Convolutional Network, which consists of a Partitioning Module (PM), a Spatio-Temporal Graph Convolution (ST-GC) module and a Joint Prototype Loss (JPL) function, which facilitates the capture of local relationships between joints.
Wen-Peng Wang, Guo-Long Qin, Xi-Zhan Gao et al.· International Journal of Mac...· 0 citations
The results indicate that skeleton-rendered pose images provide a practical accuracy-efficiency trade-off for structured physical training action recognition.
Shuo Wang, Hao Lu, Ye Zhuang et al.· Discover Applied Sciences· 0 citations
Comprehensive ablation studies reveal that the architectural innovations and the stability protocol contribute complementarily, and that BPA-STGCN achieves the best accuracy–stability trade-off among all tested configurations.