Breaking the Code on Broken Tablets: The Learning Challenge for Annotated Cuneiform Script in Normalized 2D and 3D Datasets

  title={Breaking the Code on Broken Tablets: The Learning Challenge for Annotated Cuneiform Script in Normalized 2D and 3D Datasets},
  author={Hubert Mara and Bartosz Bogacz},
  journal={2019 International Conference on Document Analysis and Recognition (ICDAR)},
  • H. Mara, B. Bogacz
  • Published 1 September 2019
  • Computer Science
  • 2019 International Conference on Document Analysis and Recognition (ICDAR)
The number of known cuneiform tablets is assumed to be in the hundreds of thousands. The Hilprecht Archive Online contains 1977 high-resolution 3D scans of tablets. The online cuneiform database CDLI catalogs metadata for more than 100.000 tablets. While both are accessible publicly, large-scale machine learning and pattern recognition on cuneiform tablets remain elusive. The data is only accessible by searching web pages, the tablet identifiers between collections are inconsistent, and the 3D… Expand
Period Classification of 3D Cuneiform Tablets with Geometric Neural Networks
  • B. Bogacz, H. Mara
  • Computer Science
  • 2020 17th International Conference on Frontiers in Handwriting Recognition (ICFHR)
  • 2020
This work combines and adapt the convolution operation of SplineNet with the pooling from PointNet++ to predict the time- period of a tablet directly on basis of its mesh representation, and reaches a classification accuracy of 84 % on 4 time-period classes. Expand
Metadata schema and ontology for capturing and processing of 3D cultural heritage objects
Motivated by the increased use of 3D acquisition of objects by cultural heritage institutions, we were investigating ontologies and metadata schemes for the acquisition process to provide detailsExpand


Web-Based Scientific Exploration and Analysis of 3D Scanned Cuneiform Datasets for Collaborative Research
The paper shows that increasing the availability of 3D scanned archaeological data through a web-based interface can provide significant scientific value while at the same time finding a trade-off between copyright induced restrictions and scientific usability. Expand
Feature Descriptors for Spotting 3D Characters on Triangular Meshes
  • B. Bogacz, H. Mara
  • Computer Science
  • 2018 16th International Conference on Frontiers in Handwriting Recognition (ICFHR)
  • 2018
A novel approach adapting relevant image descriptor methods to process irregular grids for robust symbol-spotting and transcription of ancient scripts by providing a qualitative analysis of the spotting results showing that a query mesh is reliably found on the document mesh. Expand
Retrieving Cuneiform Structures in a Segmentation-free Word Spotting Framework
This paper presents an approach for automatically retrieving cuneiform wedge constellations from digitized cunesiform tablet collections and shows compelling results could be achieved in the qualitative and quantitative evaluation on a challenging benchmark consisting of 3D-scanned cuniform tablets. Expand
Annotating a Low-Resource Language with LLOD Technology: Sumerian Morphology and Syntax
The project’s main goal is to build a pipeline for machine translation and annotation of Sumerian Ur III administrative texts, and represents the first attempt to develop an integrative infrastructure for the annotation of morphology and syntax on the basis of RDF technologies and LLOD resources. Expand
Vectorization of 3D-Characters by Integral Invariant Filtering of High-Resolution Triangular Meshes
  • H. Mara, S. Krömker
  • Computer Science
  • 2013 12th International Conference on Document Analysis and Recognition
  • 2013
A filtering algorithm on 2D-manifolds using Multi-Scale Integral Invariants (MSII) to detect characters within a high-dimensional feature space to utilize the full potential of the acquired 3D-data. Expand
Towards a Linked Open Data Edition of Sumerian Corpora
The LLOD edition of a linguistically annotated corpus of Sumerian, as well as its linking with lexical resources, repositories of annotation terminology, and the museum collections in which the artifacts bearing these texts are kept are described. Expand
New Visualization Techniques for Cuneiform Texts and Sealings
This contribution presents insights into the ongoing evolution of new visualization techniques relevant for the fields of archaeology and, particularly, Assyriology. We focus on systems and methodsExpand
Made in the humanities: Dual integral invariants for efficient edge detection
  • H. Mara
  • Computer Science
  • it Inf. Technol.
  • 2016
A fast, efficient adaption for edge detection in raster images is introduced as an alternative to the Canny edge detector algorithm and other complex algorithms using first and second order derivative filters. Expand
Line Drawings vs. Curvature Shading: Scientific Illustration of Range Scanned Artefacts
The question if line drawings could generally be replaced by a detail-shaded view, which highlights relevant features, but still conveys an objective plastic impression as well is discussed. Expand
Efficient algorithms for local and global accessibility shading
The idea of surface accessibility is extended to include “global accessibility” which measures the ability of a spherical probe to enter a structure from outside as well as to fit locally on the surface. Expand