Skip to content
This is a multi-lingual OCR which basically isolated the OCR used in E2E-MLT
Python Shell
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.

A Multi-Lingual OCR using CTC

This repository contains only the OCR unit that is used in this repository.

Please use Anaconda or miniconda for installation.

To run this model, you would need the following steps:

  • conda env create -f environment.yml
  • wget
  • conda activate ocr


simply run:

  • bash
  • conda activate ocr

make sure that you have a GPU

now you have 2 choices.

Run the OCR on images present in input_data and save the output in output_data:


Please Note: the format in which the recognition result is saved is:



if your image name is: img_1.jpg, and your recognition result is: hello_world.

The output image name would be: img_1_hello_world.png.

how to run images present in some random <input_image_path> and store output in some random <output_image_path>?

to do this instead of bash run: python -input-path=your_random_image_path -output_path=your_random_image_path


  title={E2E-MLT-an unconstrained end-to-end method for multi-language scene text},
  author={Bu{\v{s}}ta, Michal and Patel, Yash and Matas, Jiri},
  journal={arXiv preprint arXiv:1801.09919},

About me: Go to

You can’t perform that action at this time.