SImpHAR: Advancing impedance-based human activity recognition using 3D simulation and text-to-motion models

2025-07-08

AI Summary

Bioimpedance sensing holds promise for fine-grained human activity recognition (HAR), but its practical deployment is hindered by severe scarcity of labeled training data. Method: This paper proposes a simulation-driven weakly supervised learning framework comprising: (1) the first bioimpedance signal synthesis pipeline integrating soft-body physics modeling with shortest-path resistance estimation; and (2) a two-stage decoupled training strategy that avoids label alignment, leveraging 3D human mesh representations, text-to-motion generation, and photorealistic data augmentation. Contribution/Results: We achieve the first controllable, high-fidelity bioimpedance signal synthesis, effectively alleviating the labeling bottleneck. Evaluated on our newly established ImpAct dataset and two public benchmarks, our method achieves up to 22.3% improvement in accuracy and 21.8% in macro-F1 score, while substantially expanding the scope of recognizable activities.

Technology Category

• Machine Learning: Bio-inspired Learning
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Web Mining and Content Analysis: Web data generation and simulation

Abstract

Human Activity Recognition (HAR) with wearable sensors is essential for applications in healthcare, fitness, and human-computer interaction. Bio-impedance sensing offers unique advantages for fine-grained motion capture but remains underutilized due to the scarcity of labeled data. We introduce SImpHAR, a novel framework addressing this limitation through two core contributions. First, we propose a simulation pipeline that generates realistic bio-impedance signals from 3D human meshes using shortest-path estimation, soft-body physics, and text-to-motion generation serving as a digital twin for data augmentation. Second, we design a two-stage training strategy with decoupled approach that enables broader activity coverage without requiring label-aligned synthetic data. We evaluate SImpHAR on our collected ImpAct dataset and two public benchmarks, showing consistent improvements over state-of-the-art methods, with gains of up to 22.3% and 21.8%, in terms of accuracy and macro F1 score, respectively. Our results highlight the promise of simulation-driven augmentation and modular training for impedance-based HAR.

Problem

• Addressing scarcity of labeled bio-impedance data for HAR
• Generating realistic bio-impedance signals via 3D simulation
• Improving activity recognition accuracy with modular training

Innovation

• Generates bio-impedance signals using 3D simulation
• Employs text-to-motion models for data augmentation
• Uses decoupled two-stage training for broader coverage