no code implementations • 13 Nov 2015 • Ehsan Hosseini-Asl, Angshuman Guha
In this paper, we propose a new text recognition model based on measuring the visual similarity of text and predicting the content of unlabeled texts.
Spelling Correction