HeiCuBeDa Hilprecht - Heidelberg Cuneiform Benchmark Dataset for the Hilprecht Collectiondoi:10.11588/data/IE8CCNheiDATA2019-06-062Mara, Hubert, 2019, "HeiCuBeDa Hilprecht - Heidelberg Cuneiform Benchmark Dataset for the Hilprecht Collection", https://doi.org/10.11588/data/IE8CCN, heiDATA, V2HeiCuBeDa Hilprecht - Heidelberg Cuneiform Benchmark Dataset for the Hilprecht Collectiondoi:10.11588/data/IE8CCNMara, HubertBayer, Paul VictorHubert MaraBartosz Bogacz2019-03-11Heidelberg, GermanyGigaMesh Software FrameworkheiDATAMara, Hubert2019-02-25Arts and HumanitiesComputer and Information ScienceThe number of known cuneiform tablets is assumed to be in the hundreds of thousands. A fraction has been published by printing photographs and manual tracings in books, which is collected by the online Cuneiform Digital Library Initiative (CDLI) catalog including some of these images and providing metadata for more than 100.000 tablets. While 3D-acquisition of tablets is the most modern way for their documentation, the number of 3D-datasets is relatively small and often not openly accessible. However, the Hilprecht Archive Online (HAO) provides 1977 high-resolution 3D scans of tablets under an Open Access license. While both the HAO and the CDLI are accessible publicly, large-scale machine learning and pattern recognition on cuneiform tablets remains elusive, because the data is only accessible by navigating web pages, the tablet identifiers between collections are inconsistent, and the 3D data is unprepared and challenging for automated processing. We enable large-scale analysis of cuneiform tablets by this HeiCuBeda for Hilprecht assembly, which is a cross-referenced benchmark dataset of processed cuneiform tablets: (i) frontally aligned 3D tablets with pre-computed high-dimensional surface features, (ii) six-views raster images for off-the-shelf image processing, and (iii) metadata, transcriptions, and transliterations, for a subset of 707 tablets, for learning alignment between 3D data, image and linguistic expression. This is the first dataset of its kind, and of its size, in cuneiform research. This benchmark dataset is prepared for ease-of-use and immediate availability for computational researches, lowering the barrier to experiment and apply standard methods of analysis. A script in Python is provided to retrieve and compute an updated JSON database of the CDLI’s metadata and raster images. Up-to-date code and meta-data are also available at <a href="https://gitlab.com/fcgl/releases/-/tree/master/mara_icdar_2019">https://gitlab.com/fcgl/releases/-/tree/master/mara_icdar_2019</a>.2018-07-242018-08-222019-03-012019-03-11Cuneiform tablets3D Measurement dataFurther Identifiers of the persons involved:<p>
<p>
<ul>
<li>Hubert Mara: ORCID: <a href="https://orcid.org/0000-0002-2004-4153">https://orcid.org/0000-0002-2004-4153</a>, Wikidata: <a href="https://www.wikidata.org/wiki/Q97924674">https://www.wikidata.org/wiki/Q97924674</a></li>
<li>Bartosz Bogacz: ORCID: <a href="https://orcid.org/0000-0002-8323-5694">https://orcid.org/0000-0002-8323-5694</a>, Wikidata: <a href="https://www.wikidata.org/wiki/Q102869220">https://www.wikidata.org/wiki/Q102869220</a></li>
<li>Paul Victor Bayer: ORCID: <a href="https://orcid.org/0000-0003-1528-5531">https://orcid.org/0000-0003-1528-5531</a></li>
</ul>Hilprecht Sammlung, Jena, Germany, <a href="https://hilprecht.mpiwg-berlin.mpg.de/
">https://hilprecht.mpiwg-berlin.mpg.de/</a><p/>
Cuneiform Digital Library Initiative (CDLI)
<a href="https://cdli.ucla.edu/">https://cdli.ucla.edu/</a>Heidelberg Cuneiform 3D Database (HeiCu3Da) for the Hilprecht Collection: <a href="https://doi.org/10.11588/heidicon.hilprecht">https://doi.org/10.11588/heidicon.hilprecht</a>H. Mara and B. Bogacz, "Breaking the Code on Broken Tablets: The Learning Challenge for Annotated Cuneiform Script in Normalized 2D and 3D Datasets," 2019 International Conference on Document Analysis and Recognition (ICDAR), Sydney, NSW, Australia, 2019, pp. 148-153.https://doi.org/10.1109/ICDAR.2019.00032H. Mara and B. Bogacz, "Breaking the Code on Broken Tablets: The Learning Challenge for Annotated Cuneiform Script in Normalized 2D and 3D Datasets," 2019 International Conference on Document Analysis and Recognition (ICDAR), Sydney, NSW, Australia, 2019, pp. 148-153.Bartosz Bogacz and Hubert Mara: Period Classification of 3D Cuneiform Tablets with Geometric Neural Networks. In: Proceedings of the 17th International Conference on Frontiers of Handwriting Recognition (ICFHR). Dortmund, Germany 2020.https://doi.org/10.1109/ICFHR2020.2020.00053Bartosz Bogacz and Hubert Mara: Period Classification of 3D Cuneiform Tablets with Geometric Neural Networks. In: Proceedings of the 17th International Conference on Frontiers of Handwriting Recognition (ICFHR). Dortmund, Germany 2020.GigaMesh and Gilgamesh - 3D Multiscale Integral Invariant Cuneiform Character Extractionhttps://doi.org/10.2312/VAST/VAST10/131-138GigaMesh and Gilgamesh - 3D Multiscale Integral Invariant Cuneiform Character ExtractionMulti-Scale Integral Invariants for Robust Character Extraction from Irregular Polygon Mesh Data10.11588/heidok.00013890Multi-Scale Integral Invariants for Robust Character Extraction from Irregular Polygon Mesh DataHeiCuBeDa_00_Supplementary_Documentation.pdfSupplementary Documentation about the contents of the HeiCuBeDa and HeiCu3Da bundles.application/pdfHeiCuBeDa_01_Logo_1977.pdfLogo for the HeiCuBeDa Hilprecht dataset consisting of 1977 cuneiform tablets.application/pdfHeiCuBeDa_A1_Images_Sideviews_MSII_Filter.zipA complete set of six side views for each of the 1977 3D-datasets using the MSII filter response to highlight surface details i.e. cuneiform script and sealings. Recommended for learning tasks. The images are stored as PNGs.application/zipHeiCuBeDa_A2_Images_Sideviews_VirtualLight.zipComplete set of eight side views of the 3D-models rendering using a virtual light source and a metallic surface to mimic the illumination setup of photographs. The images are stored as PNGs.application/zipHeiCuBeDa_B_Hilprecht_Database_240121.jsonapplication/jsonHeiCuBeDa_B_Scrape_CDLI_240121.pytext/x-pythonHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part01.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part02.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part03.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part04.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part05.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part06.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part07.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part08.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part09.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part10.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part11.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part12.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part13.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part14.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part15.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part16.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part17.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part18.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part19.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_C_3DData_with_MSII_and_FunctionValue_part20.zipStanford Polygon (PLY) files including the feature vectors computed using the volume based integral invariant.application/zipHeiCuBeDa_D_MSII_filter_results_surface_integration_part01.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part02.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part03.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part04.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part05.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part06.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part07.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part08.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part09.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part10.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part11.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part12.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part13.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part14.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part15.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part16.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part17.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part18.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part19.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tarHeiCuBeDa_D_MSII_filter_results_surface_integration_part20.tarMulti-scale surface based integral invariants as GNU Octave MAT file compressed with bzip2. One vector per line. First column is the vertex index.application/x-tar