Achieving Robustness of Mesh Watermarking Techniques Toward Mesh Geometry and Topology-Invariant Operations

Eminent usage of three-dimensional (3D) media on the Internet nowadays brings the concern related to their security. Watermarking is a branch of security mechanism that protects these 3D models from malicious users, and it also detects and localizes the attacks. While talking about attacks to 3D media, there are a number of mesh processing operations which do not alter the actual geometry or topology of the mesh. These operations are not considered as attack, and the mesh watermarking process should be able to resist such operations. These operations include mesh translation, mesh scaling, mesh rotation, and vertex reordering. This paper discusses different approaches to make the mesh watermarking operations invariant to the above routine operations. The approaches have also been implemented to 3D meshes.