OCR-ReportLab-Notebooks

OCR-ReportLab is a collection of Colab notebooks designed to perform Optical Character Recognition (OCR) on images and generate DOCX or PDF documents containing both the original image and the extracted text. It supports multiple state-of-the-art vision-language models for experimentation and practical use.

All the notebooks are here : https://huggingface.co/prithivMLmods/OCR-ReportLab-Notebooks/tree/main

Notebooks

You can launch and run the following notebooks directly in Google Colab:

Nanonets OCR: Open in Colab
Monkey OCR: Open in Colab
OCRFlux 3B: Open in Colab
Typhoon OCR: Open in Colab
and more ...

Features

Extracts text from input images using various OCR models
Embeds the image and extracted text into DOCX or PDF formats
Designed for quick deployment via Google Colab

Supported Models

The repository currently supports the following OCR implementations:

Nanonets OCR
Monkey OCR
OCRFlux 3B
Typhoon OCR 3B
and more ...

Installation

No installation is required. Simply click on the links above to run the notebooks in Google Colab. Make sure to upload your image file(s) when prompted and follow the instructions in the notebook.

Other Images

OCR

Caption

Dependencies

The notebooks are built using:

Python
PyTorch
Hugging Face Transformers
ReportLab
Gradio (for UI)
(Qwen2.5-VL based)

All dependencies are automatically installed in the Colab environment.

Github

OCR-ReportLab-Notebooks : https://github.com/PRITHIVSAKTHIUR/OCR-ReportLab-Notebooks

Author

Created and maintained by PRITHIVSAKTHIUR