OCR Text Extraction with Byaldi and Qwen2-VL

This project implements Optical Character Recognition (OCR) using the ColPali implementation of the Byaldi library and Hugging Face's Qwen2-VL model. The extracted text from images (containing Hindi and English) is displayed through a simple web application built using Streamlit. Users can search for specific keywords within the extracted text.

Project Overview

The project involves:

• OCR Implementation: Using ColPali's Byaldi library and Hugging Face transformers (Qwen2-VL model) to extract text from images.
• Web Application: A Streamlit web interface that allows users to upload images, view extracted text, and search within the text.
• Deployment: The web application is deployed on platforms like Streamlit Sharing or Hugging Face Spaces, making it accessible via a public URL.

Features

• OCR Extraction: Extract text from images in Hindi and English using the Byaldi + Qwen2-VL model.
• Keyword Search: Search within the extracted text for specific keywords, with matching results highlighted.
• User-Friendly Interface: Upload and process images via a simple and intuitive Streamlit web interface.

Requirements

• Python 3.8 or higher
• Install dependencies using the provided requirements.txt

Installation

1. Clone the repository:

   git clone https://github.com/Ayantanu2002/Image_OCR.git
   cd Image_OCR

2. Set up the virtual environment:

   python -m venv .venv
   source .venv/bin/activate  # On Windows use .venv\Scripts\activate

3. Install the required dependencies:

   pip install -r requirements.txt

Running the Application

Once the installation is complete, you can start the Streamlit app:

streamlit run app.py