A small snake game built with MediaPipe + OpenCV + Python. Control the snake head using your finger in front of the camera. The live camera feed serves as the game canvas for an immersive CV/HCI experience.

---

• Camera feed as the game background/canvas
• Finger control via hand landmarks (MediaPipe Hands)
• Lightweight, runs fully offline
• Suitable for computer vision & HCI demos/assignments

• Python 3.8+ (recommended 3.9 or newer)
• OS: Windows (verified). macOS/Linux should also work in principle
• Dependencies:
  • opencv-python
  • mediapipe
  • numpy
• A working webcam

1. Clone or download this repository
2. Create and activate a virtual environment (recommended)
   • Windows PowerShell:
     • python -m venv .venv
     • ./.venv/Scripts/Activate.ps1
   • macOS/Linux (bash/zsh):
     • python3 -m venv .venv
     • source .venv/bin/activate
3. Install dependencies
   • If requirements.txt exists:
     • pip install -r requirements.txt
   • Or install manually:
     • pip install opencv-python mediapipe numpy

• Ensure your webcam is accessible and permission is granted
• From the project root, run:
  • python main.py
• For best results on first run, use good lighting and a simple background

• Move your index finger (or the designated finger) in front of the camera to steer the snake head
• Keep your hand clearly visible within the frame
• Depending on implementation, mechanics may include: speed vs. motion amplitude, collision detection, scoring rules, etc.

• Prefer bright, uniform lighting and a simple background
• Avoid strong reflections and occlusions
• If detection jitters, consider adding smoothing/filtering in code

• main.py: Entry point. Camera capture, MediaPipe detection, game loop
• game/: Game logic (snake body, collision, scoring)
• utils/: Utilities (coordinate mapping, smoothing)
• assets/: Optional resources

Actual files may differ; please refer to the code.

• Webcam cannot open
  • Check if another app is using the camera; verify capture index cv2.VideoCapture(0) (try 1, 2)
  • Allow camera access in system privacy settings
• Unstable hand detection
  • Improve lighting, simplify background, move hand closer; adjust MediaPipe parameters if needed
• Missing dependency errors
  • Install with pip install opencv-python mediapipe numpy
• Low FPS
  • Lower the frame resolution; optimize drawing/processing; disable unnecessary visualizations

• Image processing is local only; nothing is uploaded externally
• Use responsibly and inform subjects when applicable

• No explicit license set. If you plan to open-source, consider adding a common license (MIT/Apache-2.0, etc.)

• Issues/PRs are welcome to improve gesture stability, game mechanics, UI/UX, and performance

• MediaPipe Hands for efficient hand landmark detection
• OpenCV for image capture and processing