Review of Computational Epigraphy
Vishal Kumar
TL;DR
This review addresses a key problem in archaeology: extracting text, interpreting meaning, and attributing inscriptions from stone artifacts in a way that minimizes damage and subjectivity. It surveys transliteration and attribution efforts powered by computational imaging, rule-based methods, and various supervised and semi-supervised learning approaches across diverse scripts, including the undeciphered Indus script and Cuneiform. Notable contributions include rule-based transliteration for Cuneiform, 3D CAD-based tablet reconstruction (Gigamesh), and NLP/ML methods for predicting origin, period, and missing text, with extensive use of CNNs, SVMs, LSTMs, Transformers, and GANs. The findings suggest that computational methods significantly enhance epigraphical workflows and can yield robust insights where traditional methods struggle, signaling a shift toward ML-supported epigraphy as a default approach while emphasizing collaboration with domain experts.
Abstract
Computational Epigraphy refers to the process of extracting text from stone inscription, transliteration, interpretation, and attribution with the aid of computational methods. Traditional epigraphy methods are time consuming, and tend to damage the stone inscriptions while extracting text. Additionally, interpretation and attribution are subjective and can vary between different epigraphers. However, using modern computation methods can not only be used to extract text, but also interpret and attribute the text in a robust way. We survey and document the existing computational methods that aid in the above-mentioned tasks in epigraphy.