💪 AI Push-Up Counter (MediaPipe)

A real-time push-up counter using Computer Vision. The script analyzes the user's pose via webcam, tracks body landmarks, and automatically counts repetitions based on the vertical distance between the elbow and shoulder.

---

🚀 Features

• Real-time Pose Detection: Uses Google's MediaPipe library to track 33 body landmarks.
• Automatic Repetition Counting: Logic based on the Y-coordinate relationship between the elbow and shoulder.
• Visual Feedback: * Displays a real-time "skeleton" (landmarks and connections).
  • HUD showing the current count and state (UP/DOWN).
  • Dynamic image overlay (reward effect) triggered upon successful completion of a push-up.
• Efficient Logic: Simple state-machine approach to prevent double-counting.

---

🛠️ How It Works

1. Preprocessing: The frame is captured via OpenCV, flipped for a mirror effect, and converted from BGR to RGB (required by MediaPipe).
2. Landmark Extraction: The system identifies the LEFT_SHOULDER and LEFT_ELBOW points.
3. Overlay System: When a repetition is counted, a timer (show_effect) is activated to display a custom image on the screen for 20 frames.

---

📋 Requirements

• Python 3.9+
• Webcam
• Libraries:

  pip install opencv-python mediapipe