🕹️ Pose-Controlled Gesture Detection using MediaPipe

A Python-based computer vision system that uses MediaPipe Pose to recognize body gestures like jumping, crouching, hand-joining, and horizontal movement to control games like Subway Surfers or Temple Run without touching a keyboard!

---

🎮 Use Case: Play Games with Your Body!

This project is designed to be extended to play endless runner games using your body:

Game	Gesture	Action Triggered
Subway Surfers	Jump	Character jumps
	Crouch	Character rolls
	Move Left/Right	Character dodges
Temple Run	Hands Joined	Start or pause the game

💡 With some automation via pyautogui, gestures can be mapped to keystrokes (e.g., jump = up arrow, crouch = down arrow).

---

📦 Features

• ✅ Real-time pose tracking using MediaPipe
• 👐 Hand gesture detection (joined / not joined)
• ↔️ Horizontal movement detection (left / center / right)
• 🦵 Posture classification (jump / stand / crouch)
• 📸 Webcam-based gesture recognition
• 🧠 Easy integration with automation tools

---

🧰 Tech Stack

• Python 🐍
• OpenCV 🎥
• MediaPipe 📐
• Matplotlib 📊
• PyAutoGUI (optional for automation) ⌨️

---

🔧 Installation

1. Clone the repository:

   git clone https://github.com/yourusername/pose-gesture-detection.git
   cd pose-gesture-detection

2. Install required packages:

     pip install opencv-python mediapipe matplotlib pyautogui

3. Run the Program

     python main.py

   This will:

   • Open your webcam

   • Detect pose and classify gestures

   • Optionally simulate key presses using gestures

---

📁 Project Structure

.
├── subwaySurfers.py       # Main script to run the real-time system
├── README.md              # Project documentation

---

🙌 Credits

Built using:

• MediaPipe Pose

• OpenCV

• PyAutoGUI