EMPress: Practical Hand Gesture Classification with Wrist-Mounted EMG and Pressure Sensing
summary
Wrist-mounted system combines four pairs of EMG electrodes and four pressure sensors (FSRs), runs an SVM with RBF kernel, and is evaluated with 12 participants performing 15 gestures ten times each. EMG alone averages ~86% accuracy, pressure outperforms EMG on wrist rotations by ~10%, EMG outperforms pressure on finger flexions by ~5%, and the combined sensor set reaches 96% 10-fold cross-validated accuracy. Wet electrodes plus foam remove pressure artefacts to isolate the complementary signals.
AI-written summary · about
chapters (4)
related talks
11mWrist2Finger: Sensing Fingertip Force for Force-Aware Hand Interaction with a Ring-Watch Wearable
20mAdvancing Hand Gesture Recognition with High Resolution Electrical Impedance Tomography
13mSparseEMG: Computational Design of Sparse EMG Layouts for Sensing Gestures
06mDesigning and Evaluating Hand-to-Hand Gestures with Dual Commodity Wrist-Worn Devices
06mWR-Hand: Wearable Armband can Track User's Hand
14mLeveraging Dual-Observable Input for Fine-Grained Thumb Interaction Using Forearm EMG
19mBackHand: Sensing Hand Gestures via Back of the Hand
17mEchoFlex: Hand Gesture Recognition using Ultrasound Imaging