"Photo Semantic Finder: Replicate Google Photos' aiming to replicate its functionality of searching images using natural language queries. With this application, users can upload their images and search for them using descriptive captions such as "car", "child playing", or "birthdays".
-
Upload Images: Users can upload their images through the web interface.
-
Caption Generation: The uploaded images are passed through the BLIP Image Captioning model to generate descriptive captions.
-
Search by Caption: Users can search for images using natural language queries. The application retrieves images whose captions match the query.
-
Semantic Image Search: Utilizes state-of-the-art natural language processing models like Salesforce's BLIP Image Captioning Large to generate descriptive captions for uploaded images.
-
Streamlit Web Application: The frontend of the application is built using Streamlit, providing an intuitive user interface for uploading images and querying them using natural language.
-
Uploading and Preview
Preview image
-
Grid Interface
Preview image
-
Searching using description
Preview image
sol:
-
others
Preview image
-
Streamlit: Used for building the web application frontend.
-
Hugging Face Transformers: Leveraged the Salesforce/blip-image-captioning-large model for generating image captions.
-
Python: The backend and scripting language used for the application development.
Download the model from Hugging Face or else from here and then extract it in 'Models' folder
-
Clone the repository:
git clone https://github.com/your-username/photo-semantic-finder.git
-
Install dependencies:
pip install -r requirements.txt
-
Run the Streamlit app:
streamlit run photo_Semantics.py
Contributions are welcome! Feel free to open issues for feature requests, bug fixes, or general improvements. Pull requests are also appreciated.
This project is licensed under the MIT License - see the LICENSE file for details.