CNN based cuneiform sign detection learned from annotated 3D renderings and mapped photographs with illumination augmentation
Motivated by the challenges of the Digital Ancient Near Eastern Studies (DANES) community, we develop digital tools for processing cuneiform script being a 3D script imprinted into clay tablets used for more than three millennia and at least eight major languages. It consists of thousands of charact...
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Dokumenttyp: | Article (Journal) Kapitel/Artikel |
| Sprache: | Englisch |
| Veröffentlicht: |
22 Aug 2023
|
| In: |
Arxiv
Year: 2023, Pages: 1-9 |
| DOI: | 10.48550/arXiv.2308.11277 |
| Online-Zugang: | Verlag, kostenfrei, Volltext: https://doi.org/10.48550/arXiv.2308.11277 Verlag, kostenfrei, Volltext: http://arxiv.org/abs/2308.11277 
|
| Verfasserangaben: | Ernst Stötzner, Timo Homburg, Hubert Mara |
MARC
| LEADER | 00000caa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 1858234727 | ||
| 003 | DE-627 | ||
| 005 | 20240307055155.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 230830s2023 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.48550/arXiv.2308.11277 |2 doi | |
| 035 | |a (DE-627)1858234727 | ||
| 035 | |a (DE-599)KXP1858234727 | ||
| 035 | |a (OCoLC)1425212496 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 28 |2 sdnb | ||
| 100 | 1 | |a Stötzner, Ernst |e VerfasserIn |0 (DE-588)1300462698 |0 (DE-627)1858235316 |4 aut | |
| 245 | 1 | 0 | |a CNN based cuneiform sign detection learned from annotated 3D renderings and mapped photographs with illumination augmentation |c Ernst Stötzner, Timo Homburg, Hubert Mara |
| 264 | 1 | |c 22 Aug 2023 | |
| 300 | |a 9 | ||
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Gesehen am 30.08.2023 | ||
| 520 | |a Motivated by the challenges of the Digital Ancient Near Eastern Studies (DANES) community, we develop digital tools for processing cuneiform script being a 3D script imprinted into clay tablets used for more than three millennia and at least eight major languages. It consists of thousands of characters that have changed over time and space. Photographs are the most common representations usable for machine learning, while ink drawings are prone to interpretation. Best suited 3D datasets that are becoming available. We created and used the HeiCuBeDa and MaiCuBeDa datasets, which consist of around 500 annotated tablets. For our novel OCR-like approach to mixed image data, we provide an additional mapping tool for transferring annotations between 3D renderings and photographs. Our sign localization uses a RepPoints detector to predict the locations of characters as bounding boxes. We use image data from GigaMesh's MSII (curvature, see https://gigamesh.eu) based rendering, Phong-shaded 3D models, and photographs as well as illumination augmentation. The results show that using rendered 3D images for sign detection performs better than other work on photographs. In addition, our approach gives reasonably good results for photographs only, while it is best used for mixed datasets. More importantly, the Phong renderings, and especially the MSII renderings, improve the results on photographs, which is the largest dataset on a global scale. | ||
| 650 | 4 | |a Computer Science - Artificial Intelligence | |
| 650 | 4 | |a Computer Science - Computer Vision and Pattern Recognition | |
| 650 | 4 | |a Computer Science - Machine Learning | |
| 700 | 1 | |a Homburg, Timo |e VerfasserIn |0 (DE-588)1300461551 |0 (DE-627)1858221382 |4 aut | |
| 700 | 1 | |a Mara, Hubert |d 1975- |e VerfasserIn |0 (DE-588)1028750749 |0 (DE-627)731565584 |0 (DE-576)376225076 |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t Arxiv |d Ithaca, NY : Cornell University, 1991 |g (2023) vom: Aug., Artikel-ID 2308.11277, Seite 1-9 |h Online-Ressource |w (DE-627)509006531 |w (DE-600)2225896-6 |w (DE-576)28130436X |7 nnas |a CNN based cuneiform sign detection learned from annotated 3D renderings and mapped photographs with illumination augmentation |
| 773 | 1 | 8 | |g year:2023 |g month:08 |g elocationid:2308.11277 |g pages:1-9 |g extent:9 |a CNN based cuneiform sign detection learned from annotated 3D renderings and mapped photographs with illumination augmentation |
| 787 | 0 | 8 | |i Forschungsdaten |a Mara, Hubert, 1975 - |t MaiCuBeDa Hilprecht - Mainz Cuneiform Benchmark Dataset for the Hilprecht Collection |d Heidelberg : Universität, 2023 |h 1 Online-Ressource (11 Files) |w (DE-627)1858221188 |
| 856 | 4 | 0 | |u https://doi.org/10.48550/arXiv.2308.11277 |x Verlag |x Resolving-System |z kostenfrei |3 Volltext |
| 856 | 4 | 0 | |u http://arxiv.org/abs/2308.11277 |x Verlag |z kostenfrei |3 Volltext |
| 951 | |a AR | ||
| 992 | |a 20230830 | ||
| 993 | |a Article | ||
| 994 | |a 2023 | ||
| 998 | |g 1028750749 |a Mara, Hubert |m 1028750749:Mara, Hubert |d 110000 |e 110000PM1028750749 |k 0/110000/ |p 3 |y j | ||
| 999 | |a KXP-PPN1858234727 |e 4371018459 | ||
| BIB | |a Y | ||
| JSO | |a {"id":{"eki":["1858234727"],"doi":["10.48550/arXiv.2308.11277"]},"origin":[{"dateIssuedDisp":"22 Aug 2023","dateIssuedKey":"2023"}],"name":{"displayForm":["Ernst Stötzner, Timo Homburg, Hubert Mara"]},"relHost":[{"physDesc":[{"extent":"Online-Ressource"}],"origin":[{"dateIssuedKey":"1991","publisher":"Cornell University ; Arxiv.org","dateIssuedDisp":"1991-","publisherPlace":"Ithaca, NY ; [Erscheinungsort nicht ermittelbar]"}],"id":{"zdb":["2225896-6"],"eki":["509006531"]},"disp":"CNN based cuneiform sign detection learned from annotated 3D renderings and mapped photographs with illumination augmentationArxiv","note":["Gesehen am 28.05.2024"],"type":{"bibl":"edited-book","media":"Online-Ressource"},"recId":"509006531","language":["eng"],"pubHistory":["1991 -"],"titleAlt":[{"title":"Arxiv.org"},{"title":"Arxiv.org e-print archive"},{"title":"Arxiv e-print archive"},{"title":"De.arxiv.org"}],"part":{"year":"2023","pages":"1-9","text":"(2023) vom: Aug., Artikel-ID 2308.11277, Seite 1-9","extent":"9"},"title":[{"title":"Arxiv","title_sort":"Arxiv"}]}],"physDesc":[{"extent":"9 S."}],"title":[{"title_sort":"CNN based cuneiform sign detection learned from annotated 3D renderings and mapped photographs with illumination augmentation","title":"CNN based cuneiform sign detection learned from annotated 3D renderings and mapped photographs with illumination augmentation"}],"person":[{"given":"Ernst","family":"Stötzner","role":"aut","roleDisplay":"VerfasserIn","display":"Stötzner, Ernst"},{"family":"Homburg","given":"Timo","roleDisplay":"VerfasserIn","display":"Homburg, Timo","role":"aut"},{"display":"Mara, Hubert","roleDisplay":"VerfasserIn","role":"aut","family":"Mara","given":"Hubert"}],"recId":"1858234727","language":["eng"],"type":{"bibl":"chapter","media":"Online-Ressource"},"note":["Gesehen am 30.08.2023"]} | ||
| SRT | |a STOETZNERECNNBASEDCU2220 | ||