Skip to content

Gesture recognition using RPI Pico and Edge Impulse machine learning, can be used as a general purpose controller by mapping gestures to perform actions or shortcuts using a python script.

Notifications You must be signed in to change notification settings

yosefahab/GestureGlove

Repository files navigation

Gesture Glove

ML Gesture recognition using edge impulse & RPI pico. Detects gestures and performs shortcuts accordingly using a python script.

Fun fact: EdgeImpulse mentioned GestureGlove on linkedin on #PiDay

Requirements

  • Cmake
  • Pico SDK
  • python
  • pyserial pip3 install pyserial
  • pyautogui pip3 install pyautogui

Steps

1. connect your pico while holding the bootsel button

2. run the following:

```
mkdir build
&& cd build
&& cmake ..
&& make
```

3. copy build/gGlove.uf2 to RPI-RP2 when its connected

Connections

  • PICO 3v3 -> MPU6050 VCC
  • PICO GND -> MPU6050 GND
  • PICO GP4 -> MPU6050 SDA
  • PICO GP5 -> MPU6050 SCL
  • PICO GP0 -> HC-05 TX
  • PICO GP1 -> HC-05 RX
  • PICO VBUS -> HC-05 VCC
  • PICO GND -> HC-05 GND

Bluetooth_Reciever

Python script to recieve data from the Pico and perform shortcuts saved in config.json file.
If no custom configuration is passed in command arguments, the script uses the default configuration.

To run :
python gGloveReciever.py <COM_NUMBER> <CONFIG_OBJECT(optional)>

NOTE:

  • THIS BRANCH USES BLUETOOTH NOT USB
  • THIS PYTHON SCRIPT IS INTENDED FOR WINDOWS, IT HAS NOT BEEN TESTED WITH OTHER PLATFORMS.

About

Gesture recognition using RPI Pico and Edge Impulse machine learning, can be used as a general purpose controller by mapping gestures to perform actions or shortcuts using a python script.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages