# Optical Text Recognition in Nepali and Bengali: A Transformer-based Approach

arXiv: https://arxiv.org/abs/2404.02375 · HTML: https://mlsanity.com/paper/2404.02375/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 13 references that are in the map.
- First posted on arXiv 3 Apr 2024.

## Builds on (most cited first, 9 of 9)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Unsupervised Cross-lingual Representation Learning at Scale](https://mlsanity.com/paper/1911.02116/) (2019)
- [Visual Transformers: Token-based Image Representation and Processing for Computer Vision](https://mlsanity.com/paper/2006.03677/) (2020)
- [TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models](https://mlsanity.com/paper/2109.10282/) (2021)
- [BanglaWriting: A multi-purpose offline Bangla handwriting dataset](https://mlsanity.com/paper/2011.07499/) (2020)
- [OCR Improves Machine Translation for Low-Resource Languages](https://mlsanity.com/paper/2202.13274/) (2022)
- [An Improved Feature Descriptor for Recognition of Handwritten Bangla Alphabet](https://mlsanity.com/paper/1501.05497/) (2015)
- [End-to-End Optical Character Recognition for Bengali Handwritten Words](https://mlsanity.com/paper/2105.04020/) (2021)
- [A BLSTM Network for Printed Bengali OCR System with High Accuracy](https://mlsanity.com/paper/1908.08674/) (2019)

## Built on by (most cited first, 1 of 1)

- [Deciphering the Underserved: Benchmarking LLM OCR for Low-Resource Scripts](https://mlsanity.com/paper/2412.16119/) (2024)

## Abstract (arXiv)

Efforts on the research and development of OCR systems for Low-Resource Languages are relatively new. Low-resource languages have little training data available for training Machine Translation systems or other systems. Even though a vast amount of text has been digitized and made available on the internet the text is still in PDF and Image format, which are not instantly accessible. This paper discusses text recognition for two scripts: Bengali and Nepali; there are about 300 and 40 million Bengali and Nepali speakers respectively. In this study, using encoder-decoder transformers, a model was developed, and its efficacy was assessed using a collection of optical text images, both handwritten and printed. The results signify that the suggested technique corresponds with current approaches and achieves high precision in recognizing text in Bengali and Nepali. This study can pave the way for the advanced and accessible study of linguistics in South East Asia.