G3CN: Gaussian Topology Refinement Gated Graph Convolutional Network for Skeleton-Based Action Recognition

2025/09/09 by Haiqing Ren, Ren, Haiqing, Heng Fan +8
Computer Science · #Advanced Graph Neural Networks #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Human Pose and Action Recognition

paper · pdf · doi:10.48550/arxiv.2509.07335

openalex publication_date 2025/09/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Graph Convolutional Networks (GCNs) have proven to be highly effective for skeleton-based action recognition, primarily due to their ability to leverage graph topology for feature aggregation, a key factor in extracting meaningful representations. However, despite their success, GCNs often struggle to effectively distinguish between ambiguous actions, revealing limitations in the representation of learned topological and spatial features. To address this challenge, we propose a novel approach, Gaussian Topology Refinement Gated Graph Convolution (G³CN), to address the challenge of distinguishing ambiguous actions in skeleton-based action recognition. G³CN incorporates a Gaussian filter to refine the skeleton topology graph, improving the representation of ambiguous actions. Additionally, Gated Recurrent Units (GRUs) are integrated into the GCN framework to enhance information propagation between skeleton points. Our method shows strong generalization across various GCN backbones. Extensive experiments on NTU RGB+D, NTU RGB+D 120, and NW-UCLA benchmarks demonstrate that G³CN effectively improves action recognition, particularly for ambiguous samples.

Citations

• Learning Discriminative Representations for Skeleton Based Action Recognition
• Graph Contrastive Learning for Skeleton-based Action Recognition
• Leveraging Spatio-Temporal Dependency for Skeleton-Based Action Recognition
• Hierarchically Decomposed Graph Convolutional Networks for Skeleton-Based Action Recognition
• IGFormer: Interaction Graph Transformer for Skeleton-based Human Interaction Recognition
• Towards To-a-T Spatio-Temporal Focus for Skeleton-Based Action Recognition
• Contrastive Learning from Extremely Augmented Skeleton Sequences for Self-supervised Action Recognition
• Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action Recognition
• Multi-Scale Spatial Temporal Graph Convolutional Network for Skeleton-Based Action Recognition
• Revisiting Skeleton-based Action Recognition
• Spatio-Temporal Inception Graph Convolutional Networks for Skeleton-Based Action Recognition
• Dynamic GCN: Context-enriched Topology Learning for Skeleton-based Action Recognition
• Disentangling and Unifying Graph Convolutions for Skeleton-Based Action Recognition
• Actional-Structural Graph Convolutional Networks for Skeleton-based Action Recognition
• Semantics-Guided Neural Networks for Efficient Skeleton-Based Human Action Recognition
• Graph Edge Convolutional Neural Networks for Skeleton Based Action Recognition
• Spatial Temporal Graph Convolutional Networks for Skeleton-Based Action Recognition
• Skeleton-Based Action Recognition Using Spatio-Temporal LSTM Network with Trust Gates
• Semi-Supervised Classification with Graph Convolutional Networks
• NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysis
• Deep Residual Learning for Image Recognition
• Convolutional Networks on Graphs for Learning Molecular Fingerprints
• Deep Convolutional Networks on Graph-Structured Data
• Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation
• Cross-view Action Modeling, Learning and Recognition

Related

• Channel-wise Topology Refinement Graph Convolution for Skeleton-Based Action Recognition
• Topology-aware Convolutional Neural Network for Efficient Skeleton-based Action Recognition
• SiT-MLP: A Simple MLP with Point-wise Topology Feature Learning for Skeleton-based Action Recognition
• Topological Symmetry Enhanced Graph Convolution for Skeleton-Based Action Recognition
• Dynamic GCN: Context-enriched Topology Learning for Skeleton-based Action Recognition