Motion Matters: Motion-guided Modulation Network for Skeleton-based Micro-Action Recognition

2025-07-29

AI Summary

To address insufficient modeling of subtle motion variations in micro-action recognition, this paper proposes a Motion-Guided Modulation Network (MMN) that explicitly captures dynamic distinctions via dual-path motion cue modulation—operating at both skeletal and frame levels. Key contributions include: (1) a Motion-Sensitive Skeleton Modulation module (MSM) that injects motion awareness at the joint level; (2) a Temporal Modulation module (MTM) to enhance discriminability of inter-frame sub-pixel displacements; and (3) a multi-scale motion consistency learning strategy that jointly optimizes local and global motion representations. By integrating skeleton sequence modeling, spatiotemporal attention, and feature modulation mechanisms, MMN achieves fine-grained capture of sub-pixel dynamic changes. Extensive experiments demonstrate state-of-the-art performance on Micro-Action 52 and iMiGUE benchmarks, validating the critical efficacy of motion-guided modulation for micro-action recognition.

Technology Category

• Computer Vision: Motion & Tracking
• Machine Learning: Multimodal Learning
• Planning, Routing, and Scheduling: Activity and Plan Recognition

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Web Mining and Content Analysis: Models for Web evolution

Abstract

Micro-Actions (MAs) are an important form of non-verbal communication in social interactions, with potential applications in human emotional analysis. However, existing methods in Micro-Action Recognition often overlook the inherent subtle changes in MAs, which limits the accuracy of distinguishing MAs with subtle changes. To address this issue, we present a novel Motion-guided Modulation Network (MMN) that implicitly captures and modulates subtle motion cues to enhance spatial-temporal representation learning. Specifically, we introduce a Motion-guided Skeletal Modulation module (MSM) to inject motion cues at the skeletal level, acting as a control signal to guide spatial representation modeling. In parallel, we design a Motion-guided Temporal Modulation module (MTM) to incorporate motion information at the frame level, facilitating the modeling of holistic motion patterns in micro-actions. Finally, we propose a motion consistency learning strategy to aggregate the motion cues from multi-scale features for micro-action classification. Experimental results on the Micro-Action 52 and iMiGUE datasets demonstrate that MMN achieves state-of-the-art performance in skeleton-based micro-action recognition, underscoring the importance of explicitly modeling subtle motion cues. The code will be available at https://github.com/momiji-bit/MMN.

Problem

• Recognizing micro-actions with subtle motion changes
• Enhancing spatial-temporal representation in skeleton-based recognition
• Improving accuracy in micro-action classification via motion cues

Innovation

• Motion-guided Modulation Network for subtle cues
• Skeletal and frame level motion modulation
• Multi-scale motion consistency learning strategy