Learning Adaptive Node Selection with External Attention for Human Interaction Recognition

2025-07-05

AI Summary

Existing GCN-based methods for modeling human–human interaction typically treat individuals as isolated graph nodes, neglecting inherent inter-person dependencies; moreover, predefined adjacency matrices lack adaptability to capture action-specific and context-sensitive joint interactions. To address these limitations, we propose a prior-free dynamic node selection mechanism: an Adaptive Temporal Node Amplitude Computation (AT-NAC) module identifies kinematically salient motion nodes, while an External Attention (EA) mechanism models cross-subject dynamic interaction dependencies. Our approach integrates GCN with spatial motion magnitude analysis, learnable threshold-based node filtering, and temporal weighting—enabling precise perception of salient motion patterns and efficient interaction modeling. Evaluated on multiple benchmark datasets, the method achieves state-of-the-art performance, significantly enhancing flexibility and representational capacity for relational modeling in complex interactive scenarios.

Technology Category

• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Humans and AI: Interaction Techniques and Devices
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Most GCN-based methods model interacting individuals as independent graphs, neglecting their inherent inter-dependencies. Although recent approaches utilize predefined interaction adjacency matrices to integrate participants, these matrices fail to adaptively capture the dynamic and context-specific joint interactions across different actions. In this paper, we propose the Active Node Selection with External Attention Network (ASEA), an innovative approach that dynamically captures interaction relationships without predefined assumptions. Our method models each participant individually using a GCN to capture intra-personal relationships, facilitating a detailed representation of their actions. To identify the most relevant nodes for interaction modeling, we introduce the Adaptive Temporal Node Amplitude Calculation (AT-NAC) module, which estimates global node activity by combining spatial motion magnitude with adaptive temporal weighting, thereby highlighting salient motion patterns while reducing irrelevant or redundant information. A learnable threshold, regularized to prevent extreme variations, is defined to selectively identify the most informative nodes for interaction modeling. To capture interactions, we design the External Attention (EA) module to operate on active nodes, effectively modeling the interaction dynamics and semantic relationships between individuals. Extensive evaluations show that our method captures interaction relationships more effectively and flexibly, achieving state-of-the-art performance.

Problem

• Model dynamic human interactions without predefined adjacency matrices
• Adaptively select relevant nodes for interaction recognition
• Capture semantic relationships between individuals using active nodes

Innovation

• Dynamic interaction capture without predefined assumptions
• Adaptive node selection with spatial-temporal weighting
• External Attention module for modeling interaction dynamics