A downloadable tool for Windows, macOS, and Linux

Custom gesture recognition. Leap commander encodes static hand gestures as shell commands to help make your desktop experience more efficient. Hand gestures can be stored in database for future use.

Leap Commander uses machine learning to decide which command your making so if your gestures aren't being recognized you can simply add more data and update your model.

Install instructions

Leap Commander is written entirely in Python. To run it, simply decompress the LeapCommander.zip source and run

>>> python main.py

in the terminal in the source folder. You'll need to download the following python packages as dependencies:

- Scikit-learn

- Numpy

- PyQt4

- h5py

Once you've launched Leap Commander, click on File >> New Database to start a new command database. Record commands with the "New Command" button and delete commands with the "Delete Command" button. You can add data for each gesture by making your desired hand gesture above the Leap motion and hitting the "Teach Command" button. Make sure the click on the command you'd like to teach in the command table before trying to teach it.

Before you can execute commands make sure to update the internal model used by Leap Commander by hitting the "Update Model" button. Once your database is made and you've got an active model, interpret hand gestures by hitting the "Execute" button while gesturing above the Leap Motion.


LeapCommander.zip 3 MB

Leave a comment

Log in with itch.io to leave a comment.