Content-Based Indexing and Retrieval of Cultural Heritage Data: An Integrated Approach to Documentation with Application to the EROS
Bibliographic record
Abstract
Over the last few decades, Cultural Heritage documentation has been characterized by the massive use of digital media. Recently, the use of three-dimensional scanner technologies has provided us with the opportunity to obtain an unambiguous body of information characterizing the three-dimensional shapes of the artefacts. These vast repositories are now structured in databases, for easy access. Such databases contain not only the artefacts, but also relevant information such as restoration reports, data regarding quantitative analysis, chemical formulae, etc. It follows that storing such information is not enough. Rather, it should be indexed in order to be searched and retrieved easily and rapidly. In addition, the data should be preserved as technologies evolve over time, in order to ensure long-term preservation and access. This paper presents a framework for indexing and retrieval of 2D and 3D Cultural Heritage data. In our approach, novel archiving and indexing techniques, developed by the National Research Council of Canada, are employed. We present the results as applied to the EROS (EuropeanResearch Open System) Database of the C2RMF. This database consists of an impressive collection of scientific and technical data about paintings and artefacts found in all the museums of France. Our results indicate that our content-based approaches are able to accurately index and retrieve diverse images and 3D objects, based on the artefacts as well as their fragments. That is, using for example a fragment of a picture, we are able to retrieve the correct image even in conditions where lighting, orientation and the surroundings of the reference are different. The content-based retrieval system is also able to retrieve different views of the same object, e.g. of a Chalcidian amphora. In addition, our approach is able to find groups of similar images or objects, such as white figurines from the same period or 3D scans of an Anadyomene Venus.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.005 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".