KinEMbed: Decoding Kinematics from Electromyography via Cross-Modal Contrastive Learning
2026-07-06 • Machine Learning
Machine Learning
AI summaryⓘ
The authors developed a new method called KinEMbed to better predict hand movements from muscle signals measured on the skin. Unlike past methods that mostly focused on classifying hand gestures, their approach works to continuously estimate joint angles. They use two interconnected neural network encoders to learn shared features from both muscle signals and hand movement data. Tested on a dataset including people with and without limb differences, KinEMbed performed better than other common methods, especially for tricky thumb movements. This is an early step toward improving wearable devices for prosthetics and rehabilitation using continuous movement estimation.
surface electromyography (EMG)hand kinematicsregressioncontrastive learningembeddingsjoint anglesprosthetic controlrepresentation learningNinaPro datasetdegrees of articulation
Authors
Sofia Gilardini, Chenfei Ma, Kianoush Nazarpour
Abstract
Decoding hand kinematics from surface electromyography (EMG) is a core challenge in wearable biosignal processing with clinical relevance for prosthetic control and motor rehabilitation. Most representation learning approaches for EMG focus on discrete gesture classification, and few focus on continuous regression. We present KinEMbed, a cross-modal contrastive learning framework for hand kinematics regression that jointly trains dual encoders -- one for windowed EMG features and one for kinematic (joint angle) targets. The resulting embeddings inherit the geometric structure of the kinematic space without requiring kinematic signals at inference time. Evaluating on the NinaPro DB8 dataset that includes both able-bodied users and subjects with limb difference (N=11), KinEMbed outperforms PCA, PLS, autoencoder and contrastive (CEBRA) baselines on held-out sessions, with largest gains on the most challenging thumb degrees of articulation. We position this work as a first step toward contrastive representation learning for regression of hand kinematics from structured wearable biosignals.