Reconstruction of Three-Dimensional Objects through Matching of Their Parts
IEEE Transactions on Pattern Analysis and Machine Intelligence
Virtual Archaeologist: Assembling the Past
IEEE Computer Graphics and Applications
A Flexible Similarity Measure for 3D Shapes Recognition
IEEE Transactions on Pattern Analysis and Machine Intelligence
Reassembling fractured objects by geometric matching
ACM SIGGRAPH 2006 Papers
Integral invariants for robust geometry processing
Computer Aided Geometric Design
Using Non Local Features for 3D Shape Grouping
SSPR & SPR '08 Proceedings of the 2008 Joint IAPR International Workshop on Structural, Syntactic, and Statistical Pattern Recognition
IEEE Transactions on Systems, Man, and Cybernetics, Part B: Cybernetics
| Hi-index | 0.00 |
In this paper we propose a method for characterizing the surface of fractured pieces which come from old and complete original objects. Natural fractured pieces are difficult to segment due to the fact that the faces and edges are not well defined. For this reason, standard local feature based approaches are clearly inefficient to make an efficient segmentation. Our segmentation procedure is based on the Cone Curvature (CC) concept applied over the original dense models which provide standard scanner modeling tools. This CC based-method allows us to explore the surface from multiple neighborhood levels and to find a compact segmentation solution which characterizes different parts of the piece. A wide experimentation has been carried out on a set of old fractured pieces belonging to the remains of roman sculptures.