Python Hand Tracking

A simple webcam hand-tracking app built with OpenCV and MediaPipe.

What it does

• Opens your webcam
• Detects hands in real time
• Draws hand landmarks on the video feed
• Swaps left and right labels to match the mirrored preview

Requirements

• Windows
• Python installed
• Visual Studio Code installed
• A webcam

First-time setup in VS Code

1. Open VS Code.
2. Choose File > Open Folder....
3. If you have not downloaded the project yet, clone it from GitHub:

git clone <your-repo-url>

4. Open the python-hand-tracking folder in VS Code.
5. Open the integrated terminal in VS Code.

Create and activate the virtual environment

If the .venv folder already exists, activate it:

Set-ExecutionPolicy -Scope Process -ExecutionPolicy RemoteSigned
.\.venv\Scripts\Activate.ps1

If .venv does not exist, create it first:

py -m venv .venv
.\.venv\Scripts\Activate.ps1

Install dependencies

Run this inside the activated virtual environment:

pip install opencv-python mediapipe

Run the program

python main.py

If python does not work in your terminal, try:

py main.py

How to quit

• Press q in the video window, or
• Click the X button on the window title bar

Notes

• The first run downloads the MediaPipe hand landmarker model automatically.
• Keep the webcam connected and allow camera access if Windows asks for permission.
• If the window opens but stays black, close it, wait a moment, and run the program again.