Related Experiment Video
Updated: Aug 5, 2026
08:15
Capturing Dynamic Finger Gesturing with High-resolution Surface Electromyography and Computer Vision
Published on: March 28, 2025
A Co-Located sEMG-pFMG Dataset for Hand Gesture Recognition Under Varying Arm-Position Conditions
Shen Zhang1, Hao Zhou1, Rayane Tchantchane1
1Applied Mechatronics and Biomedical Engineering Research (AMBER) Group, University of Wollongong, Wollongong, NSW 2522, Australia.
Sensors (Basel, Switzerland)
|July 28, 2026
Summary
This study introduces a new dataset for hand gesture recognition (HGR) using wearable sensors, combining surface electromyography (sEMG) and pressure-based force myography (pFMG) signals. The dataset supports developing robust HGR algorithms under varied arm postures and movements.
Area of Science:
- Biomedical Engineering
- Human-Computer Interaction
- Wearable Technology
Background:
- Hand gesture recognition (HGR) using wearable sensors faces challenges from variations in arm posture, motion, and muscle activation.
- Existing datasets often lack multi-modal sensing or realistic dynamic conditions.
Purpose of the Study:
- To present a comprehensive, multi-modal dataset for HGR research.
- To facilitate the systematic investigation of sensing strategies under realistic conditions.
- To provide a benchmark for evaluating HGR algorithms and human-machine interfaces.
Main Methods:
- Developed a custom armband for co-located sEMG and pFMG signal acquisition.
- Collected data across three subsets: static posture, multiple static postures, and combined static/dynamic postures.
- Validated hardware signal quality (SNR) and performed baseline gesture recognition experiments.
Main Results:
- Demonstrated stable and reliable signal acquisition with good signal-to-noise ratios.
- Provided qualitative insights into modality-specific and condition-dependent signal characteristics.
- Established reproducible baseline performance using conventional machine learning classifiers.
Conclusions:
- The presented dataset is a valuable resource for HGR algorithm development and evaluation.
- Enables research into multi-modal sensing strategies for wearable human-machine interfaces.
- Supports advancing robust gesture recognition under diverse and dynamic arm conditions.
