SparseEMG: Computational Design of Sparse EMG Layouts for Sensing Gestures
summary
Computational tool generates sparse forearm EMG layouts from a chosen gesture set and ML parameters and predicts accuracy; evaluated across six datasets (28 selection/classifier combos) and validated with eight participants and two hardware setups (Olimex EMG shield, 8‑channel EMG band). Empirical search finds Permutation Importance paired with Random Forest gives the best tradeoff, cutting electrodes by 53.5% while real‑world tests show only a 2–4% accuracy drop. The tool supports 50+ gestures and produces printable stencils for per‑user placement.
AI-written summary · about
chapters (6)
- 0:00Introduction to SparseEMG
- 1:48Tool workflow and features
- 5:18Electrode selection analysis
- 7:24Cross-user validation results
- 8:49Real-world application validation
- 9:32Summary and future work
related talks
05mA Feature Adaptive Learning Method for High-Density sEMG-Based Gesture Recognition
17mEMPress: Practical Hand Gesture Classification with Wrist-Mounted EMG and Pressure Sensing
20mEarFieldSensing: A Novel In-Ear Electric Field Sensing to Enrich Wearable Gesture Input through ...
06mDesigning and Evaluating Hand-to-Hand Gestures with Dual Commodity Wrist-Worn Devices
17mEchoFlex: Hand Gesture Recognition using Ultrasound Imaging
10mDextrEMS: Achieving Dexterity in Electrical Muscle Stimulation by Combining it with Brakes
34mThe hand prosthesis that learns from the user
18mM.Gesture: An Acceleration-Based Gesture Authoring System on Multiple Handheld and Wearable Devices