Hand Tracking using MediaPipe

This Python script utilizes the MediaPipe library to perform hand tracking in real-time using your webcam. It detects hand landmarks and draws connections between them, allowing you to visualize hand movements.

Sample Output

---

Prerequisites

Before running the script, ensure you have the following installed:

• Python (version 3.7 or higher)
• OpenCV (opencv-python package)
• MediaPipe (mediapipe package)

You can install the required packages using pip:

pip install opencv-python mediapipe==0.10.18

Note: Newer versions of mediapipe may cause errors with this project.

Usage

1. Clone this repository to your local machine.
2. Navigate to the directory containing the script.
3. Run the script using Python:

python hand_tracking.py

1. A window will pop up displaying your webcam feed with hand tracking overlay.
2. Press 'q' to exit the program.

Code Explanation

• The script captures video input from your webcam (cv2.VideoCapture).
• It utilizes MediaPipe's Hands model to detect hand landmarks in each frame.
• Detected landmarks are used to draw circles at specific key points on the hand (e.g., fingertips, base).
• Hand connections are drawn between landmarks to visualize hand structure.
• The script calculates and displays frames per second (FPS) to monitor performance.
• Pressing 'q' quits the application.

Mediapipe Hand Landmarks

---