Similarity reasoning over collections of 3D artifacts: the GRAVITATE example

The creation of an increasing number of high-fidelity 3D models of objects and environments has opened new opportunities to study the past, by giving access to plenty of representations of artefacts close to their original form. Besides their 3D geometry, objects are possibly equipped with photometric data, chemical properties and digitized card catalogue information on provenance, classification, and full-text archaeological descriptions from various sources. In the GRAVITATE project, we are analysing to what extent existing similarity techniques are able to answer to the fragment similarity problem, discussing what is currently missing and what is necessary to be further developed. In this contribution, we present the criteria adopted for shape similarity reasoning and the solutions adopted in the current version of the geometric search engine of the GRAVITATE project. Since the concept of similarity is not unique, the geometric search engine takes in consideration multiple aspects of each model, mostly local (like color, thickness ...) but not only those (like its global shape), and can do searches keeping in count one or more aspects at once. The way these aspects, or properties, are encoded in the search engine and, based on these last, how models are considered similar are the two main challenges tackled in search engine design and dataset exploration. Our aim is to develop a flexible and multi-modal reasoning able to support cross-correlation and searching across collections. Currently we are envisaging to drive similarity reasoning by the overall fragment size, thickness, material texture, shape continuity, colour, decorations and overall shape. First results are promising and a first prototype of the search engine is currently available.