Microbit RGB lamp controlled with machine learning HandPose

This project uses a micro:bit that communicate via Bluetooth with computer. Ml5.js and p5.js libraries are used to program an app with Machine Learning.

Handpose is a machine learning model that allows for palm detection and hand-skeleton finger tracking in the browser. It can detect a maximum of one hand at a time and provides 21 3D hand keypoints that describe important locations on the palm and fingers.

By holding your hand in front of the web camera the program detects it and shows it on screen. I used the position of tip och indexfinger to calculate when it points on the colored squares in app.

The RGB led on micro:bit changes color depending on whitch square is "touched"

The video shows how it works

I 3D printed a flower and a vase for the RGB Led and connected it to the micro:bit large pins 0, 1 and 2.

I made a program with p5.js and Ml5.js that connects the computer to the micro:bit over Bluetooth.

You can see the code on my GitHub and try it out on my GitHub page.

The micro:bit is programmed in micro:bit MakeCode editor.

To view the code in a larger image visit my GitHub.