A web application that uses AI vision to identify different types of Barbie dolls from photos. Simply take a picture of a Barbie doll, and the app will analyze it to tell you which Barbie it is.
- 📸 Real-time camera access using device's camera
- 🤖 AI-powered Barbie identification
- 📱 Mobile-friendly responsive design
- 🖼️ Image preview before submission
- ✨ Clean and intuitive user interface
-
Take a Picture:
- Open the app in your browser
- Grant camera permissions when prompted
- Position your code in frame
- Click the "Take Picture" button
- Review the captured image
-
AI Analysis:
- The captured image is securely sent to our AI vision model
- The model analyzes the image to identify the product code
- Results are returned within seconds
-
View Results:
- See detailed information about the identified Barbie
- Results include the type of Barbie and other relevant details
- Take another picture to scan a different Barbie
- Frontend: Next.js 13+ with React
- Styling: Tailwind CSS
- Camera Interface: Web APIs (getUserMedia)
- AI Vision: Advanced image recognition model
-
Clone the repository:
git clone https://github.com/yourusername/barbie-scanner.git
-
Install dependencies:
cd barbie-scanner npm install -
Install Ollama https://ollama.com/
-
Run llama 3.2-vision model
ollama run llama3.2-vision
-
Run the development server:
npm run dev:https
-
Open https://localhost:3000 in your browser
docker build -t barbie-scanner .- Modern web browser with camera access
- Camera permissions enabled
- JavaScript enabled
- Internet connection for AI model access
- Images are processed securely and not stored permanently
- Camera access is only used when taking pictures
- No personal data is collected or stored
Feel free to contribute to this project by submitting issues or pull requests. All contributions are welcome!
MIT License - feel free to use this project for your own purposes.