DeepHadad: Enhancing the Readability of Ancient Northwest Semitic Inscriptions

Simple item page
dc.contributor.authorAioanei, Andrei C.en_US
dc.contributor.authorKlein, Jonathanen_US
dc.contributor.authorKlein, Konstantin M.en_US
dc.contributor.authorHunziker-Rodewald, Regine R.en_US
dc.contributor.authorMichels, Dominik L.en_US
dc.contributor.editorCorsini, Massimilianoen_US
dc.contributor.editorFerdani, Danieleen_US
dc.contributor.editorKuijper, Arjanen_US
dc.contributor.editorKutlu, Hasanen_US
dc.date.accessioned2024-09-15T09:56:25Z
dc.date.available2024-09-15T09:56:25Z
dc.date.issued2024
dc.description.abstractWe present DeepHadad, a novel deep learning approach to improve the readability of severely damaged ancient Northwest Semitic inscriptions. By leveraging concepts of displacement maps and image-to-image translation, DeepHadad effectively recovers text from barely recognizable inscriptions, such as the one on the Hadad statue. A main challenge is the lack of pairs of well-preserved and damaged glyphs as training data since each available glyph instance has a unique shape and is not available in different states of erosion. We overcome this issue by generating synthetic training data through a simulated erosion process, on which we then train a neural network that successfully generalizes to real data. We demonstrate significant improvements in readability and historical authenticity compared to existing methods, opening new avenues for AI-assisted epigraphic analysis.en_US
dc.description.sectionheadersCapturing and Visualizing the Visible and Invisible
dc.description.seriesinformationEurographics Workshop on Graphics and Cultural Heritage
dc.identifier.doi10.2312/gch.20241242
dc.identifier.isbn978-3-03868-248-6
dc.identifier.issn2312-6124
dc.identifier.pages6 pages
dc.identifier.urihttps://doi.org/10.2312/gch.20241242
dc.identifier.urihttps://diglib.eg.org/handle/10.2312/gch20241242
dc.publisherThe Eurographics Associationen_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectCCS Concepts: Computing methodologies → Mesh geometry models; Reconstruction; Neural networks; Applied computing → Arts and humanities
dc.subjectComputing methodologies → Mesh geometry models
dc.subjectReconstruction
dc.subjectNeural networks
dc.subjectApplied computing → Arts and humanities
dc.titleDeepHadad: Enhancing the Readability of Ancient Northwest Semitic Inscriptionsen_US
Files
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
gch20241242.pdf
Size:
2.9 MB
Format:
Adobe Portable Document Format
Download
Collections
GCH 2024 - Eurographics Workshop on Graphics and Cultural Heritage