This project is an AI-powered web application that suggests outfits based on a user's wardrobe, weather conditions, and outfit preferences. The application uses deep learning models for clothing classification and background removal, and it is deployed on AWS Elastic Beanstalk.
- Upload images of your clothes and categorize them (Top, Bottom, Outerwear, etc.).
- Automatically remove backgrounds from clothing images.
- Predict the clothing category (Top, Bottom, Outerwear, etc.) using a deep learning model.
- Create outfits by selecting clothes from your virtual wardrobe.
- Get outfit suggestions based on weather data (integrated with OpenWeather API).
- Store your wardrobe in a SQLite database.
- Flask: For building the web application.
- FastAI: For clothing classification using a ResNet34 model.
- U²-Net: For background removal.
- SQLite: For storing the uploaded wardrobe items.
- OpenWeather API: For fetching weather data.
- AWS Elastic Beanstalk: For hosting the application.
- Install Python 3.10+.
- Install pip.
- Set up an account with OpenWeather API and obtain an API key.
-
Clone the repository:
git clone https://github.com/yourusername/ai-outfit.git
-
Change directory to the project folder:
cd ai-outfit -
Install the required dependencies:
pip install -r requirements.txt
-
Set up environment variables for the OpenWeather API:
export OPENWEATHER_API_KEY=your_api_key
-
Run the Flask development server:
python application.py
-
Access the application at:
http://127.0.0.1:5000/
- Navigate to the upload page to add images of clothing items.
- Use the outfit creation tool to mix and match items from your wardrobe.
- Check the weather-based outfit suggestions for daily use.
This project is licensed under the MIT License - see the LICENSE file for details.
- FastAI for the deep learning models.
- U²-Net for background removal.
- OpenWeather API for weather data.