Leap Myo

This project correlates hand landmarks from a Leap Motion Controller with EMG data from a Myo Armband.

Requirements

• Python
• Leap Motion Controller
• Myo Armband

Getting started

• Install Ultraleap Gemini
• Charge your Myo Armband and have your Leap Motion Controlller on hand

Usage

Collecting data

1. Connect your Leap Motion Controller and have your charged Myo Armband on hand
2. Run collect.py for 15 minutes, moving your hands within the frame of the Leap Motion Controller

When you quit the python process, collected data will be stored in a pandas DataFrame with each row containing a timestamp, 8 channels of EMG information and hand landmarks.

Training and testing the model

1. Run train.py to train a model on the collected data
2. With your Myo Armband on and Leap Motion Controller connected, predict hand landmarks with predict.py

Resources

• NeuroPose
• Keras
• Leap API
• Myo API