Gemma3 OCR

Project Overview

Gemma-3 OCR is a Streamlit-based application that extracts structured text from images using the Gemma-3 Vision model via Ollama. The app allows users to upload an image, process it, and receive extracted text in a structured Markdown format.

Features

• Image Upload: Users can upload PNG, JPG, or JPEG images.

• OCR Extraction: Utilizes the Gemma-3 Vision model to extract and format text.

• Clear Results: Users can clear extracted results with a button.

• User-Friendly Interface: A simple and interactive UI using Streamlit.

Installation

Installation Steps:

1. Clone the repository: git clone https://github.com/Goodnight77/gemma_ocr.git
2. Navigate to the project directory: cd gemma_ocr
3. Ensure you have Python 3 installed and the following dependencies:

Prerequisites:

  pip install streamlit ollama 

Usage

1. Navigate to the project directory: cd gemma3_OCR
2. Run the application using: streamlit run app.py
3. Upload an image and click the Extract Text button to process it.

Project Structure

  .
  ├── app.py              # Main application script
  ├── images/
  │   ├── gemma3.png      # Logo for the UI
  └── README.md           # Project documentation

Acknowledgments

• Built using Streamlit

• Powered by Ollama's Gemma-3 Vision