AI-powered car control system based on TensorFlow.js and Teachable Machine, using camera to recognize user gestures and control virtual car movement.
- 🎯 Real-time Pose Detection: Uses TensorFlow.js for real-time human pose detection
- 🚗 Virtual Car Control: Supports forward, backward, turning and other operations
- 🎮 Gesture Control: Control car movement direction through gestures
- ⚙️ Gear Shifting: Supports forward gear (D) and reverse gear (R) switching
- 📱 Responsive Design: Adapts to different screen sizes
- 🎨 Real-time Visualization: Displays pose keypoints and skeleton
Simply open the car.html file to experience
- Clone the project to local
- Use a local server to open
car.html(avoid CORS issues) - Click "Start" button to begin recognition
- Use gestures to control the car
- Forward Gesture: Car moves forward
- Stop Gesture: Car slows down and stops
- Turn Left Gesture: Car turns left
- Turn Right Gesture: Car turns right
- Spacebar: Switch between forward/reverse gear
- Frontend Framework: Native HTML5 + JavaScript
- Graphics Library: p5.js
- AI Model: TensorFlow.js + Teachable Machine Pose
- Pose Recognition: PoseNet
- Styling: CSS3
AI-Car-Pose-Control/
├── car.html # Main program file
├── cartopview.svg # Car top view icon
├── README.md # Project documentation
├── LICENSE # Open source license
└── .gitignore # Git ignore file
The project uses Google Teachable Machine for pose model training:
- Visit Teachable Machine
- Select "Pose Project"
- Train your custom poses
- Replace the model URL in the code
This project is licensed under the MIT License - see the LICENSE file for details.
Issues and Pull Requests are welcome!
For questions or suggestions, please contact through GitHub Issues.
⭐ If this project helps you, please give it a Star!
Car icon resources from: https://freesvg.org/