EMG Recordings for Pinch and Rock-Paper-Scissors Gestures from 22 Subjects

Surface electromyography (sEMG) recordings from the forearm for hand gesture recognition. The dataset contains two sets: one with 4 pinch gestures recorded from 22 subjects, and another with 3 rock-paper-scissors gestures recorded from 10 subjects. Data was collected by Elisa Donati at ETH Zurich using a Myo armband with 8 sensors at 200 Hz, with each gesture recorded 5 times per session.

Use Cases

• Train gesture classification models based on 8-channel sEMG signals.
• Benchmark real-time EMG processing algorithms for the rock-paper-scissors gestures.
• Study muscle activation patterns for pinch movements between the thumb and different fingers.
• Develop and test reservoir computing or neuromorphic systems on event-based biological signal data.

Strengths

• Includes data from 22 subjects for pinch gestures and 10 subjects for rock-paper-scissors gestures.
• Recordings are structured with 3 sessions per subject and 5 repetitions per gesture, each lasting 2 seconds.
• Specifies a 1-second relaxing phase between gestures to reduce residual muscular activation.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Elisa Donati, ETH Zurich

Collection Method

Recorded with a Myo armband composed of 8 non-invasive sEMG sensors placed around the forearm.