Image guided reconstruction of un-sampled data: a coherent filling for uncomplete cultural heritage models

Nowadays, 3D scanning is widely applied to Cultural Heritage applications. In this peculiar context, the fact that the 3D models obtained from scanning are usually uncomplete can be an issue, especially because automatic hole filling techniques can create inaccurate surfaces. In this paper, we present a method to reconstruct unsampled portions of the 3D models by inferring information about the real shape of the missing part from an image. The needed data for surface creation are extracted from a predefined pattern which is projected on the real object, in the zone where the geometry of the 3D model is missing. The procedure, which is almost completely automatic, analyzes the image in order to extract the pattern and estimate the projector position. Then, the extracted information is used to obtain a hole filling which is coherent with the real shape of the object. A series of test on real objects proves that our method is able to recover geometrical features that cannot be reconstructed using state-of-the-art methods. Consequently, it can be used to obtain complete 3D models without creating false data.