This repo contains simple project files using MediaPipe
Customized modules for hand tracking using MediaPipe is located in ./Module folder.
The required packages can be installed:
- pip install opencv-python
- pip install mediapipe
Controlling the volume of the system using hand-landmarks. To control the system volume of the Ubuntu from the python script, we will need alsaaudio.
If you are using windows, you can use pycaw and some lines of code need to be changed.
The test code can be run using:
python3 Volume_control.py
The volume of the system is mapped with the distance between two fingers. You may need to re-map according to your resulution of the webCam.
Displaying the numbers according to the fingers of the hand. MediaPipe detects the land-marks of the finger and the number is counted according the location of the landmarks on camera pixels. To display the finger images according to the results, the images are stored in Finger folder.
The test code can be run using:
python3 Finger_count.py
You can easily add for ZERO by adding hand picture in the Finger folder and only need few lines to change.