PhotoFinder Pro is a powerful application that helps you easily find your photos by searching through your albums with natural language queries. With advanced AI capabilities, PhotoFinder Pro generates descriptions for your images and allows you to retrieve them effortlessly.
- AI-powered Image Description: Automatically generate descriptions for your photos using state-of-the-art AI models.
- Natural Language Search: Search for your images using simple, natural language queries.
- Real-time Processing: View the progress of image processing in real-time.
- User-friendly Interface: Easy-to-use interface built with Streamlit.
- Python 3.9 or higher
- Streamlit
- Pillow
- google-api-python-client
- langchain-core
- langchain-community
- python-dotenv
-
Clone the Repository:
git clone [email protected]:elharchaoui/Image_Retrieval.git cd Image_Retrieval
-
Create and Activate a Virtual Environment:
python -m venv env source env/bin/activate # On Windows use `env\Scripts\activate`
-
Install the Dependencies:
pip install -r requirements.txt
-
Set Up Environment Variables: Create a
.env
file in the project root directory and add your Google API key:Google_gemini_key=YOUR_GOOGLE_API_KEY
-
Organize Your Images: Ensure your images are organized in album folders (e.g.,
album1
,album2
, etc.). -
Run the Application:
streamlit run image_retrieval.py
-
Load Images: Click the "Load Images" button to start loading and processing images from the specified album paths.
-
Search for Images: Enter a natural language query in the text input to search for images. The application will display the retrieved image based on the query.
Here's an example of how to use PhotoFinder Pro:
- Organize your images in folders named
album1
,album2
, etc. - Run the application using
streamlit run image_retrieval.py
. - Click "Load Images" to process the images.
- Enter a query like "when I was in the kitchen" to retrieve images matching the description.
This project is licensed under the MIT License. See the LICENSE file for details.
For any inquiries or issues, please contact [email protected]. Website : https://mohamedelharchaoui.com/