A fun, interactive computer vision game inspired by the classic 90’s duck hunting arcade games 🎮 — where you aim and shoot birds using a real-world object (blue pen) instead of a mouse!
This project is inspired by the nostalgic 90’s bird shooting games where a dog would bring down the birds after you shot them.
Instead of using a controller, this version uses computer vision to track a colored object (blue pen) and turn it into a real-time aiming system.
Bringing retro gaming + modern AI vision together 🚀
BIRD.mp4
- 🎯 Real-time object tracking using OpenCV (HSV color detection)
- 🔫 Aim using a blue pen
- 🐦 Moving bird with physics-like motion
- 💥 Hit detection with animation
- 📈 Score tracking system
- ⚡ Smooth gameplay with live webcam feed
- Python 🐍
- OpenCV 👁️
- CvZone
- NumPy
- Webcam captures live video
- Frame converted to HSV color space
- Blue object is detected using thresholding
- Largest contour = pen tip position
- Crosshair follows pen
- If crosshair hits bird → score increases 🎉
git clone git@github.com:your-username/bird-hunt.git
cd bird-hunt
pip install -r requirements.txt
python main.py- Move blue pen → Aim
- Hit bird → Score increases
- Press Q → Quit game
You can tweak:
lower_color = np.array([100, 150, 100])
upper_color = np.array([140, 255, 255])👉 Adjust based on lighting conditions
- 🔊 Gunshot sound effects
- 🐦 Multiple birds
- ⏱️ Timer mode (30 sec challenge)
- 🎯 Difficulty levels
- 🤖 Hand tracking instead of color detection
- 🧠 AI-based object detection (YOLO)
This project demonstrates:
- Real-time computer vision
- Interactive system design
- Game logic implementation
- Practical OpenCV skills
Osita Bharti
Robotics enthusiast | Future AI + Vision Engineer 🚀
If you like this project, give it a ⭐ on GitHub!