A Python application that captures an image from your camera and uses Ollama to get AI feedback on your outfit style. Available in two versions: a web-based interface and a console-based script.
- Python 3.9+
- Ollama installed and running locally
- Pull the
llavamodel:ollama pull llava - For web version: Flask installed (included in requirements)
- Clone or download this project.
- Create a virtual environment:
python -m venv .venv - Activate it:
source .venv/bin/activate - Install dependencies:
pip install opencv-python ollama flask rich
Run the web app for a modern, interactive experience:
python app.py- Open your browser and go to
http://127.0.0.1:5000/ - View the live camera feed on the left
- Click "Take Picture & Get Feedback" to capture and analyze
- See the Ollama style feedback displayed on the right
Run the console script for a terminal-based experience:
python main.py- The camera feed opens in a window
- Press 'c' to capture the image
- Press 'q' to quit without capturing
- Ollama feedback is printed in the terminal with rich formatting
- Live camera feed (web: in-browser, console: OpenCV window)
- One-click/image capture and AI analysis
- Structured style feedback from Ollama (Overall Impression, Color Analysis, Fit, Accessories, Suggestions)
- Friendly, encouraging AI responses
- Modern web UI with full-screen layout
- Console version with colored output and icons
- Camera Access Denied: On macOS, grant camera permissions to Python/Terminal in System Preferences > Security & Privacy > Privacy > Camera.
- Ollama Errors: Ensure Ollama is running (
ollama serve) and thellavamodel is pulled. - Web App Not Loading: Check that Flask is installed and no other app is using port 5000.