vesuvius-challenge
Unveiling the secrets of an ancient library buried by Mount Vesuvius, this Kaggle competition, supported by the Vesuvius Challenge organization, tasked participants with detecting ink from 3D X-ray scans of charred scrolls preserved in a Roman villa in Herculaneum.
Science Score: 44.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
✓CITATION.cff file
Found CITATION.cff file -
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
○Academic publication links
-
○Academic email domains
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (10.3%) to scientific vocabulary
Keywords
Repository
Unveiling the secrets of an ancient library buried by Mount Vesuvius, this Kaggle competition, supported by the Vesuvius Challenge organization, tasked participants with detecting ink from 3D X-ray scans of charred scrolls preserved in a Roman villa in Herculaneum.
Basic Info
- Host: GitHub
- Owner: association-rosia
- License: mit
- Language: Jupyter Notebook
- Default Branch: main
- Homepage: https://www.kaggle.com/competitions/vesuvius-challenge-ink-detection
- Size: 51.5 MB
Statistics
- Stars: 4
- Watchers: 0
- Forks: 0
- Open Issues: 0
- Releases: 0
Topics
Metadata Files
README.md
📜 Vesuvius Challenge - Ink Detection
This exciting competition, available on Kaggle and supported by the Vesuvius Challenge organization, aims to bring back to life an ancient library buried under the ashes of a volcano. In this competition, our mission was to detect ink from 3D X-ray scanners. Thousands of scrolls were part of a library located in a Roman villa in Herculaneum. This villa was buried by the eruption of Mount Vesuvius nearly 2000 years ago. The scrolls were charred by the heat of the volcano and are now impossible to open without damaging them.
🏆 Challenge ranking
The score of the challenge was the F0.5 score.
Our solution was in the top 10% (out of 1249 teams) with a F0.5 score equal to 0.620813 🎉.
The podium:
🥇 ryches - 0.682693
🥈 RTX23090 - 0.682443
🥉 wuyu - 0.681137
🛠️ Data processing
Tilling Method
In order to process the ultra-high definition images effectively, we employed an image tiling method. The images were divided into smaller sub-images with a size of 256 by 256 pixels to feed into the model. This approach allowed us to handle the large dataset more efficiently. Additionally, we implemented a selection criterion where we only considered tiles that contained a minimum of 5% ink pixels. By focusing on these specific tiles, we were able to concentrate our efforts on the areas most likely to contain valuable information within the ancient scrolls.
🏛️ Model architecture
📝 Citing
@misc{RebergaUrgell:2023,
Author = {Louis Reberga and Baptiste Urgell},
Title = {Vesuvius Challenge - Ink Detection},
Year = {2023},
Publisher = {GitHub},
Journal = {GitHub repository},
Howpublished = {\url{https://github.com/association-rosia/vesuvius-challenge}}
}
🛡️ License
Project is distributed under MIT License
👨🏻💻 Contributors
Louis REBERGA 
Baptiste URGELL 
Johan MONCOUTIÉ
Owner
- Name: RosIA
- Login: association-rosia
- Kind: organization
- Location: France
- Twitter: AssoRosIA
- Repositories: 1
- Profile: https://github.com/association-rosia
Citation (CITATION.cff)
cff-version: 1.2.0 message: "If you use this software, please cite it as below." authors: - family-names: "REBERGA" given-names: "Louis" - family-names: "URGELL" given-names: "Baptiste" title: "GitHub repository" publisher: "Github" year: "2023" version: 1.0 date-released: 2023-8-1 url: "https://github.com/association-rosia/vesuvius-challenge" data: "Vesuvius Challenge"
GitHub Events
Total
- Watch event: 2
Last Year
- Watch event: 2