Normal vectors and spanning tree for 3D object fingerprinting

In this paper we present a robust three-dimensional fingerprinting algorithm for verification, indexing, and identification. The core idea behind our technique is to capture the distribution of the normal vectors of a 3D triangle mesh, followed by partitioning the mesh into smaller sub-meshes. The hash vector of a 3D mesh is defined in terms of the mean values of several intervals extracted from the normal distribution and the minimum spanning tree of each sub-mesh. The experimental results on several 3D meshes prove the robustness of the proposed technique against the most common attacks including distortion-less attacks, compression, noise, smoothing, scaling, rotation as well as combinations of these attacks.