Saluja_Final thesis - Cherie Lau.pdf (21.11 MB)

Robust multilingual OCR: from Ancient Indic texts to modern Indian Street signs

Download (21.11 MB)
thesis
posted on 24.03.2021, 03:32 by Rohit Saluja
Indic Texts often contain out-of-vocabulary (OOV) words leading to significant errors in text scanned with Optical Character Recognition (OCR). We present an interactive framework for adaptively assisting Indic OCR corrections. Subsequently, we propose OCR correction methods for resource-constrained settings, such as multi-OCR consensus, plug-in classifiers, and LSTMs with a fixed delay. Modern Indian street signs and license plates pose an even tougher reading challenge. They often appear in a variety of languages, fonts, sizes, and orientations. We present the first scene text recognition results using multi-headed attention models.

History

Campus location

Australia

Principal supervisor

Mark Carman

Additional supervisor 1

Parag Chaudhuri

Additional supervisor 2

Ganesh Ramakrishnan

Year of Award

2021

Department, School or Centre

Caulfield School of IT

Additional Institution or Organisation

IITB-Monash

Course

Doctor of Philosophy

Degree Type

Doctorate

Usage metrics

Categories

Exports