Image Captioning Application Overview:

This project combines sentiment analysis with image captioning using the Salesforce BLIP model and Hugging Face's Transformers library. The chatbot analyzes text input sentiment and provides captions for uploaded images.

[INPUT PROCESSING WINDOW]

Tools Used:

Python: Backend development and machine learning. Flask: Web framework for handling backend requests. Streamlit: Tool for deploying and serving the web application. Hugging Face Transformers: Library for NLP and image captioning. HTML, CSS, JavaScript: Frontend development for user interaction.

Setup:

Install Dependencies: pip install transformers torch torchvision flask streamlit pillow

Run the Application:

Backend:

python app.py Frontend (Streamlit):

streamlit run frontend.py Accessing the Application:

Open your browser and navigate to http://localhost:8501 to interact with the chatbot and image captioning interface.

File Structure:

app.py: Flask backend for processing requests. frontend.py: Streamlit frontend for user interface. requirements.txt: List of Python dependencies.

Usage:

Sentiment Analysis:

Enter text in the chat interface and receive sentiment analysis results.

Image Captioning:

Upload an image to get automated captions generated by the BLIP model. [RESULT GENERATION WINDOW] [EXPLAINS VARIUOUS FUNCTIONBALITY OF ADDING AN IMAGE THROUGH WEBCAM, SYSTEM IMAGES AND SO ON]

Deployment:

Deployed using GRADIO for easy web deployment and sharing (Streamlit/Flask or other cloud platforms can also be used). Please find the deployed link of web application in description for testing/implementation purposes