Typefaces and Ligatures in Printed Arabic Text: A Deep Learning-Based OCR Perspective.
Omar AlhubaitiIrfan AhmadPublished in: ICDAR Workshops (2) (2023)
Keyphrases
- deep learning
- optical character recognition
- arabic text
- unsupervised learning
- scanned documents
- machine learning
- unsupervised feature learning
- character recognition
- document images
- scanned images
- mental models
- weakly supervised
- viewpoint
- deep architectures
- feature selection
- d objects
- active learning
- multiscale
- data mining