Watch Demo Video
This Python project can run alone. This program will use your camera and will process the image into landmarks. The landmarks can be used for training the hand gesture recognition model.
- Python3.9 (this is the specific version that I use on this project)
- Use the venv
Linux:
source venv/bin/activate
Windows:
.\venv\Scripts\activate
- Install the needed library from requirements.txt
pip install -r /path/to/requirements.txt
or
python3 -m pip install -r /path/to/requirements.txt
- To deactivate the venv
deactivate
Will enter record mode
In record mode, you can record currently detected hand landmarks and output it into a .csv file. The default file name is landmark.csv
. The .csv file is automatically incremented.
Return to normal mode (detecting hands only)
hand-gesture-recognition-using-mediapipe in English
hand-gesture-recognition-using-mediapipe Japanese(Original)