Improving OCR on Indic Scripts

Jun 1, 2020

We devise techniques to improve and correct OCR on documents written in Indic scripts. The methods have lead to the development of robust systems that are used to make old and new books writted in various scripts available in a digital format.

indic ocr

Publications

Sub-Word Embeddings for OCR Corrections in Highly Fusional Indic Languages
Texts in Indic Languages contain a large proportionof out-of-vocabulary (OOV) words due to frequent fusion using conjoining rules (of …
PDF Project
Sub-Word Embeddings for OCR Corrections in Highly Fusional Indic Languages
Error Detection and Correction in Indic-OCR
Optical Character Recognition (OCR) is the process of converting the document images into an editable electronic format. This has many …
Project Video
Error Detection and Correction in Indic-OCR