Alignment of Historical Handwritten Manuscripts Using Siamese Neural Network

Majeed Kassis,Jumana Nassour,Jihad El-Sana
DOI: https://doi.org/10.1109/icdar.2017.56
2017-11-01
Abstract:Historical manuscript alignment is a widely known problem in historical document analysis, and the attempt of finding the differences between manuscript editions is mainly done by hand. Today, most of the computational tools coming to assist the historians are based on word recognition or spotting. These solutions are partial at best. In this paper, we present a Siamese neural network based system, which automatically identifies whether a pair of images contain the same text without the need of recognizing the text. The user is required to annotate several pages of two manuscripts, and with the assistance of synthetically generated data and affine distortions we can align two manuscripts written by different writers, achieving strong results.
What problem does this paper attempt to address?