This is a Streamlit-based image search application that allows users to search for images based on a text query using Sentence Transformers and cosine similarity. Users can provide images by uploading a zip file, specifying a directory path, or uploading individual images.
- Upload a zip file containing images for searching.
- Specify a directory path where images are stored.
- Upload individual images for searching (default option).
- Enter a text query to find relevant images.
- Display the most relevant image and the top 10 relevant images based on cosine similarity.
-
Clone the repository:
git clone https://github.com/rediatbrook/image-search-app.git cd image-search-app
-
Install the required dependencies:
pip install -r requirements.txt
-
Run the application:
python -m streamlit run ui_test.py
- Choose one of the options to provide images: upload a zip file, specify a directory path, or upload individual images.
- Enter a text query in the input field.
- Click the "Search" button.
- The most relevant image and the top relevant images will be displayed based on the text query.
This project is licensed under the MIT License - see the LICENSE file for details.