Multiscale surface matching method for assembly or recognition of 3D archaeological objects

In archaeology, the reconstruction of ancient fragments (pottery, mosaics, marble tablets) is often a real challenge for experts studying the subject. This is due to the natural degradation of the materials (water, wind, friction) over hundreds or thousands of years. These pieces, originally inseparable, are now devoid of any visual logic of assembly. Certain visual and geometric clues enable the most obvious assemblies to be made, but this remains limited to human perception and requires more precise processing of geometric features. We propose a method for matching surfaces by mapping geometric features such as curvatures, peaks and surface normals, extracted by segmenting fragments into simple primitives (circles). This mapping method is not dependent on the rotation or translation of the fragments. We present first results on a synthetic dataset (similar to the real dataset) and conclude with the optimization process in development to deal with the real database.