EpiSAM: Character Segmentation in Challenging Stone Inscriptions
This work addresses the challenging problem of automated character segmentation in stone inscriptions for epigraphy researchers, offering a novel approach that outperforms existing methods.
EpiSAM introduces a prompt-guided transformer with a neighbor-aware strategy for character segmentation in stone inscriptions, achieving consistent improvements over baselines and strong zero-shot generalization.
Stone inscriptions are invaluable sources of historical and linguistic knowledge, yet their automated analysis remains a major challenge due to surface irregularities, erosion, and low visual contrast. Conventional document and handwriting analysis techniques fail to perform well in these scenarios. In this work, we propose character detection as a core strategy for robust inscription analysis. We introduce EpiSAM, a prompt-guided transformer framework for character segmentation in stone inscriptions. Rather than treating characters in isolation, EpiSAM employs a novel neighbor-aware strategy, explicitly predicting adjacent characters alongside the target. These contextual cues resolve boundary ambiguities, improving mask generation and enabling more accurate character segmentation. Furthermore, we expand an existing stone inscription dataset by adding dense polygonal annotations for characters, thereby enabling comprehensive research on Southeast Asian epigraphy. Experimental results show that EpiSAM achieves consistent improvements over existing baselines, while also exhibiting strong zero-shot generalization in challenging epigraphic scenarios.