GigaMesh software framework

from Wikipedia, the free encyclopedia
GigaMesh software framework

GigaMesh Logo.png
GigaMesh Screenshot 20200115 Shalmaneser III Tablet Fragment.png
Basic data

developer Forensic Computational Geometry Laboratory (FCGL), IWR , Heidelberg University
Current  version v.200529
(May 29, 2020)
operating system Linux and Windows 10
programming language C ++
License GPL
German speaking No
https://gigamesh.eu/

The GigaMesh software framework is a free and open source modular software for the display, processing and visualization of 3D data, typically recorded by means of strip light scanning or structure-from-motion .

For the analysis of archaeological objects in particular such as cuneiform tablets or ceramics , it has a multitude of functions such as roll-offs , profile sections , curvature and distance visualizations as well as export functions as raster or vector graphics . Making weathered texts legible, for example on tombstones or cuneiform tablets, or making fine surface details visible with the help of the multi-scale integral invariant (MSII) filter process is a core functionality of the software. In order to get the most reliable filter results possible, the triangular grids or polygon networks of the 3D models can be checked and adjusted. The cleaned data are also suitable for 3D printing and electronic publication in a Dataverse .

The name "GigaMesh" refers on the one hand to the processing of large 3D data sets, on the other hand it is deliberately based on Gilgamesh , a mythical Sumerian king and his heroic epic handed down on clay tablets. : 115 The logo consists of the cuneiform character ? (cascal) and can mean path, street or intersection. This means that the humanities and computer science come together. The surrounding circular shape refers to the calculation of the integral variants by means of spheres and spheres . The color red is used by the University of Heidelberg, where GigaMesh was developed.

Use and development in research projects

GigaMesh was initially developed in the context of the cuneiform texts from Assur literary content (KAL) edition of the Heidelberg Academy of Sciences . At the same time, it was used in the Corpus Vasorum Antiquorum of the Austrian Academy of Sciences to document red-figure ceramics . In addition to projects such as the multidimensional visualization and experience of cultural heritage (MUSiEKE), the framework is being expanded in a DFG and a BMBF project to contextualize and develop the corpus of the Minoan and Mycenaean seals . Analogous to the work with cuneiform tablets, there are other approaches to combining 3D image processing with methods of machine learning and used for writing in 3D, for example in the text database and dictionary of Classical Maya .

During the field campaign of the DAI in Guadalupe, near Trujillo , Honduras, comparisons were made between 3D visualizations and hand drawings in 2017. Since then, GigaMesh has been in continuous use, which has resulted in many improvements in the user interface in line with UX design . These are made available online in tutorials and are often related to the found documentation .

2018 Gigamesh was in scanning for Syria (SfS) project of the University of Leiden used to micro-CT scans of casting forms of excavation at Tell Sabi Abyad as a 3D model to reconstruct, in the Syrian civil war from the museum in al-Raqqa have disappeared. Subsequently, further CT recordings were made at the TU Delft in order to virtually extract clay tablets from clay envelopes that had not been opened for thousands of years. The SfS project won the Europa Nostra European Heritage Award in the field of research in May 2020 .

With version 190416, a version for Windows 10 was made available for the first time in the run-up to the international CAA 2019 .

The command line interface from GigaMesh is particularly suitable for the mass preparation of repositories of 3D measurement data. Almost 2,000 clay tablets from the Hilprecht collection were prepared and published digitally under CC BY licenses as a benchmark database (HeiCuBeDa) for machine learning or as an image database including metadata -containing 3D data (HeiCu3Da) . Since HeiCu3Da in an instance of EasyDB is accommodated, are a number of interfaces such. B. IIIF available. As in many other digital humanities projects, image data can be addressed directly.

On the occasion of the exhibition in the Louvre for the 100th anniversary of the CVA project, roll-offs of an aryballos from KFU Graz were calculated with GigaMesh as part of the Corpus Vasorum Antiquorum Austria in order to represent the latest methods of research into Greek vase painting . The visualizations were on display in the second half of 2019 in the showcase L'ère du numèrique et de l'imagerie scientifique .

Version 191219 supports texture mapping for the first time , which is mainly used for 3D data from photogrammetric processes. This enables the processing and, above all, the unrolling of objects that - in the field of documentation of cultural heritage and archeology - have been recorded in the widespread structure-from-motion.

The national research institute for cultural goods in Nara uses GigaMesh to unroll and document vessels and has published instructions for this. This is used for ceramics from the Jōmon period in the Togariishi Jōmon Archeological Museum in Chino .

In April 2020, the source code was published at GitLab and the license was changed from freeware to GPL . Version 200529 makes it possible for the first time to use the MSII filter in the graphical user interface to make the finest details such as fingerprints on cuneiform tablets visible.

File formats and research data infrastructures

The polygon file format is primarily supported and used to store additional information from the processing. This is not possible with the - additionally supported - Wavefront OBJ due to its specification. The marking of interpolated points and triangles by filling imperfections in the triangular grid represents meta information that z. B. should be recorded within the framework of the National Research Data Infrastructure (NFDI). Further metadata such as inventory numbers , material and hyperlinks or digital object identifiers (DOIs) can be recorded. There is also the option of calculating topological key figures that describe the quality of a 3D measurement data set.

Web links

Commons : GigaMesh Software Framework  - collection of images, videos and audio files

Individual evidence

  1. ↑ fcgl.iwr.uni-heidelberg.de
  2. ↑ gigamesh.eu
  3. ↑ a b GigaMesh Open Source Repository at GitLab. Retrieved April 10, 2020 .
  4. ↑ Paul Bayer, Susanne Lamm: More than just Ben Hur - A 3D unrolling of the Roman silver cup from Grünau , Styria . In: Forum Archaeologiae . tape 87 , no. VI , 2018, ISSN  1605-4636 .
  5. ↑ Hubert Mara, Julia Portl: Acquisition and Documentation of Vessels using High-Resolution 3D-Scanners . In: Elisabeth Trinkl (Ed.): Corpus Vasorum Antiquorum Austria . Supplement 1. Verlag der Österreichischen Akademie der Wissenschaften, 2013, ISBN 978-3-7001-7145-4 , p. 25–40 ( austriaca.at [PDF; 900 kB ; accessed on August 24, 2018]).
  6. ↑ Kurt F. de Swaaf: Chiseled secrets: Researchers decipher Jewish grave inscriptions . In: Spiegel Online . June 30, 2010 ( spiegel.de [accessed August 3, 2018]).
  7. ↑ a b Hubert Mara: Multi-Scale Integral Invariants for Robust Character Extraction from Irregular Polygon Mesh Data . Heidelberg 2012, doi : 10.11588 / heidok.00013890 .
  8. ↑ Hubert Mara, Susanne Krömker, Stefan Jakob, Bernd Breuckmann: GigaMesh and Gilgamesh - 3D Multiscale Integral Invariant Cuneiform Character Extraction . In: Proceedings of VAST International Symposium on Virtual Reality, Archeology and Cultural Heritage . Eurographics Association, 2010, ISSN  1811-864X , p. 131–138 , doi : 10.2312 / VAST / VAST10 / 131-138 ( eg.org [accessed August 3, 2018]).
  9. ↑ Hubert Mara: Visual Computing for Analysis of Sealings, Script and Fingerprints in 3D, presentation on the multi-scale integral invariant filter method with explanatory images . 2017 ( uni-koeln.de [PDF; 8.7 MB ; accessed on August 24, 2018]).
  10. ↑ heiDATA - IWR Computer Graphics Dataverse in the Heidelberg University Library. Retrieved April 16, 2019 .
  11. ↑ Regine Kleber: HEiKA - HEiKA Projects 2015. (No longer available online.) October 4, 2018, archived from the original on September 11, 2018 ; accessed on November 28, 2019 .
  12. ↑ DFG - GEPRIS - Minoan seal glyptic between corpus-like detection and 3D forensics. A multidisciplinary documentation of 900 unpublished seals from the Heraklion Archaeological Museum. Retrieved September 11, 2018 .
  13. ↑ ErKon3D - Development and contextualization of Aegean seals and seal impressions with 3D forensics. Retrieved September 11, 2018 .
  14. ↑ Bartosz Bogacz, Nikolas Papadimitriou, Diamantis Panagiotopoulos and Hubert Mara: Recovering and Visualizing Deformation in 3D Aegean Sealings . In: Proceedings of the 14th International Conference on Computer Vision Theory and Application (VISAPP) . Prague, Czech Republic 2019 ( insticc.org [accessed March 28, 2019]).
  15. ↑ Bartosz Bogacz, Felix Feldmann, Christian Prager, Hubert Mara: Visualizing Networks of Maya Glyphs by Clustering Subglyphs . In: Proceedings of the 16th Eurographics Workshop on Graphics and Cultural Heritage (GCH) . 2018, ISSN  2312-6124 , p. 105–111 , doi : 10.2312 / gch.20181346 ( eg.org [accessed February 3, 2020]).
  16. ↑ Markus Reindel , Peter Fux and Franziska Fecher: Archaeological Project Guadalupe: Report on the 2017 field campaign . In: Annual Reports . SLSA, Swiss-Liechtenstein Foundation for Archaeological Research Abroad, Zurich, Switzerland 2018, doi : 10.5167 / uzh-158145 .
  17. ^ Website of the research project Scanning for Syria at the University of Leiden. Retrieved November 27, 2019 .
  18. ↑ Dominique Ngan-Tillard: Scanning for Syria - digital book of cuneiform tablet T98-34 . June 5, 2018, doi : 10.4121 / uuid: 0bd4470b-a055-4ebd-b419-a900d3163c8a (English, 4tu.nl [PDF; 48.6 MB ]).
  19. ↑ Olivier Nieuwenhuyse, Khaled Hiatlih, Ayham al-Fakhri, Rasha Haqi, Dominique Ngan-Tillard, Hubert Mara, Katrina Burch Joosten: Focus Raqqa: Protection for the archaeological heritage of the museum of ar-Raqqa . In: Ancient World . wbg Philipp von Zabern , 2019, p. 76–83 ( wbg-wissenverbindet.de [accessed on November 27, 2019]).
  20. ↑ Seeing through clay: 4000 year old tablets in hypermodern scanners on YouTube .
  21. ↑ Unpacking a Cuneiform Tablet wrapped in a clay envelop on YouTube , cf. doi: 10.11588 / heidok.00026892 .
  22. ↑ Scanning for Syria as winner of the Europa Nostra Prize. Retrieved May 7, 2020 .
  23. ^ Website of the Scanning for Syria project at the Leiden-Delft-Erasmus Center for Global Heritage and Development . Retrieved May 15, 2020 .
  24. ↑ International Conference on Computer Applications and Quantitive Methods in Archeology, Krakow, Poland, 2019. Retrieved April 16, 2019 .
  25. ↑ Hubert Mara: HeiCuBeDa Hilprecht - Heidelberg Cuneiform Benchmark Dataset for the Hilprecht Collection . heiDATA - institutional repository for research data of Heidelberg University, June 7, 2019, doi : 10.11588 / data / IE8CCN .
  26. ↑ Hubert Mara: HeiCu3Da Hilprecht - Heidelberg Cuneiform 3D Database - Hilprecht Collection . heidICON - The Heidelberg property and multimedia database, June 7th, 2019, doi : 10.11588 / heidicon.hilprecht .
  27. ↑ Hubert Mara and Bartosz Bogacz: Breaking the Code on Broken Tablets: The Learning Challenge for Annotated Cuneiform Script in Normalized 2D and 3D Datasets . In: Proceedings of the 15th International Conference on Document Analysis and Recognition ( ICDAR ) . Sydney, Australia 2019, doi : 10.1109 / ICDAR.2019.00032 .
  28. ↑ Detail of the front of the cuneiform tablet HS 833 with the character Zeichen (cascal), referenced via pixel coordinates with origin (68,318), width 132 and height 124. Retrieved on August 5, 2020 .
  29. ^ IWR Newsroom, Contribution of visualizations to an archeological Exhibition in the Louvre Museum. July 23, 2019, accessed on January 31, 2020 .
  30. ↑ 文化 財 の 壺 第 7 号 特集: 文化 財 研究 を 進 め る 技術 を 考 え る, Issue 7, June 2019. Accessed April 10, 2020 (Japanese).
  31. ↑ Araki Minoru: Blog entry on unrolling vessels from the Jōmon period with GigaMesh. January 30, 2020, accessed April 10, 2020 (Japanese).
  32. ↑ MSII Filtering: Cuneiform Characters & Fingerprints on YouTube