Micro-DualNet: Dual-Path Spatio-Temporal Network for Micro-Action Recognition

2026-04-22

AI Summary

This work addresses the challenge of micro-action recognition, which involves intricate spatial configurations and temporal dynamics that are difficult for existing single-path spatiotemporal modeling approaches to capture effectively. To this end, the authors propose a dual-path network that models body-part entities along both spatial-temporal (ST) and temporal-spatial (TS) pathways, complemented by an entity-level adaptive routing mechanism that dynamically selects the optimal pathway for each body part. Additionally, a Mutual Action Consistency (MAC) loss function is introduced to enhance the model’s robustness to the inherent diversity of micro-actions. Integrating anatomically guided spatial partitioning, the proposed method achieves state-of-the-art performance on the iMiGUE dataset and delivers strong results on MA-52.

Technology Category

• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Computer Vision: Motion & Tracking
• Intelligent Robots: Motion and Path Planning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Micro-actions are subtle, localized movements lasting 1-3 seconds such as scratching one's head or tapping fingers. Such subtle actions are essential for social communication, ubiquitously used in natural interactions, and thus critical for fine-grained video understanding, yet remain poorly understood by current computer vision systems. We identify a fundamental challenge: micro-actions exhibit diverse spatio-temporal characteristics where some are defined by spatial configurations while others manifest through temporal dynamics. Existing methods that commit to a single spatio-temporal decomposition cannot accommodate this diversity. We propose a dual-path network that processes anatomically-grounded spatial entities through parallel Spatial-Temporal (ST) and Temporal-Spatial (TS) pathways. The ST path captures spatial configurations before modeling temporal dynamics, while the TS path inverts this order to prioritize temporal dynamics. Rather than fixed fusion, we introduce entity-level adaptive routing where each body part learns its optimal processing preference, complemented by Mutual Action Consistency (MAC) loss that enforces cross-path coherence. Extensive experiments demonstrate competitive performance on MA-52 dataset and state-of-the-art results on iMiGUE dataset. Our work reveals that architectural adaptation to the inherent complexity of micro-actions is essential for advancing fine-grained video understanding.

Problem

• micro-action recognition
• spatio-temporal characteristics
• fine-grained video understanding
• computer vision
• subtle movements

Innovation

• dual-path network
• spatio-temporal modeling
• adaptive routing
• micro-action recognition
• Mutual Action Consistency