Braille Recognition Using Deep Learning

Date
2021-08-13
Authors
Li, C
Yan, W
Supervisor
Item type
Conference Contribution
Degree name
Journal Title
Journal ISSN
Volume Title
Publisher
ACM
Abstract

Text is the media to convey and transmit information. Braille is extremely important for vision impaired people to exchange information through reading and writing. A braille translator is crucial tool for aiding people to understand braille messages. In this paper, we implement character-based braille translator using ResNet, there are three versions of ResNet we implement for braille classifiers, including ResNet-18, ResNet-34, and ResNet-50. We also implement a word-based braille detector using a novel solution called Adaptive Bezier-Curve Network (ABCNet), which is a Scene Text Recognition (STR) method for detecting word-based text in natural scenes. A comparison is present to evaluate the performance of ABCNet.

Description
Keywords
Braille recognition; Deep learning; Convolutional neural network; Natural scene text detection
Source
In 2021 4th International Conference on Control and Computer Vision (ICCCV'21). Association for Computing Machinery, New York, NY, USA, 30–35. DOI:https://doi.org/10.1145/3484274.3484280
Rights statement
© ACM, 2021. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in PUBLICATION (see Citation), (see Publisher’s Version).