Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Active Inference for Micro-Gesture Recognition: EFE-Guided Temporal Sampling and Adaptive Learning

Weijia Feng, Jingyu Yang, Ruojia Zhang, Fengtao Sun, Qian Gao, Chenyang Wang, Tongtong Su, Jia Guo, Xiaobai Li, Minglai Shao

TL;DR

An active inference-based framework for micro-gesture recognition, featuring Expected Free Energy (EFE)-guided temporal sampling and uncertainty-aware adaptive learning, offering an interpretable and scalable paradigm for temporal behavior modeling under low-resource and noisy conditions.

Abstract

Micro-gestures are subtle and transient movements triggered by unconscious neural and emotional activities, holding great potential for human-computer interaction and clinical monitoring. However, their low amplitude, short duration, and strong inter-subject variability make existing deep models prone to degradation under low-sample, noisy, and cross-subject conditions. This paper presents an active inference-based framework for micro-gesture recognition, featuring Expected Free Energy (EFE)-guided temporal sampling and uncertainty-aware adaptive learning. The model actively selects the most discriminative temporal segments under EFE guidance, enabling dynamic observation and information gain maximization. Meanwhile, sample weighting driven by predictive uncertainty mitigates the effects of label noise and distribution shift. Experiments on the SMG dataset demonstrate the effectiveness of the proposed method, achieving consistent improvements across multiple mainstream backbones. Ablation studies confirm that both the EFE-guided observation and the adaptive learning mechanism are crucial to the performance gains. This work offers an interpretable and scalable paradigm for temporal behavior modeling under low-resource and noisy conditions, with broad applicability to wearable sensing, HCI, and clinical emotion monitoring.

Active Inference for Micro-Gesture Recognition: EFE-Guided Temporal Sampling and Adaptive Learning

TL;DR

An active inference-based framework for micro-gesture recognition, featuring Expected Free Energy (EFE)-guided temporal sampling and uncertainty-aware adaptive learning, offering an interpretable and scalable paradigm for temporal behavior modeling under low-resource and noisy conditions.

Abstract

Micro-gestures are subtle and transient movements triggered by unconscious neural and emotional activities, holding great potential for human-computer interaction and clinical monitoring. However, their low amplitude, short duration, and strong inter-subject variability make existing deep models prone to degradation under low-sample, noisy, and cross-subject conditions. This paper presents an active inference-based framework for micro-gesture recognition, featuring Expected Free Energy (EFE)-guided temporal sampling and uncertainty-aware adaptive learning. The model actively selects the most discriminative temporal segments under EFE guidance, enabling dynamic observation and information gain maximization. Meanwhile, sample weighting driven by predictive uncertainty mitigates the effects of label noise and distribution shift. Experiments on the SMG dataset demonstrate the effectiveness of the proposed method, achieving consistent improvements across multiple mainstream backbones. Ablation studies confirm that both the EFE-guided observation and the adaptive learning mechanism are crucial to the performance gains. This work offers an interpretable and scalable paradigm for temporal behavior modeling under low-resource and noisy conditions, with broad applicability to wearable sensing, HCI, and clinical emotion monitoring.
Paper Structure (20 sections, 13 equations, 7 figures, 4 tables)

This paper contains 20 sections, 13 equations, 7 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Micro-gesture Recognition
  4. Active Inference
  5. Adaptive learning of uncertainty perception
  6. Problem Formulation
  7. Methodology
  8. Framework Overview
  9. EFE-Guided Temporal Selection
  10. EFE-Guided Spatial Selection
  11. Uncertainty-Aware Augmentation
  12. Experiments
  13. Experimental Settings
  14. Comparative Experiments
  15. Convergence Analysis
  16. ...and 5 more sections

Figures (7)

  • Figure 1: Overview of existing methods and UAAI
  • Figure 2: Overall Framework. The framework enhances micro-gesture recognition performance through EFE-based temporal and spatial selection and uncertainty-aware augmentation under the active inference mechanism.
  • Figure 3: Active Observation Visualization
  • Figure 4: Convergence curves of UAAI under different training epochs. The upper row shows accuracy curves, and the lower row shows corresponding loss curves. The model converges stably after around 40 epochs.
  • Figure 5: Convergence curves with and without the UMIX.
  • ...and 2 more figures