A robust and blind 3D watermarking algorithm using multiresolution adaptive parameterization of surface

A new watermarking scheme based on classification of vertices is proposed for 3D mesh models. The multiresolution adaptive parameterization of surface (MAPS) approach is used to categorize the vertices of the model into two levels: the vertices at the coarse level are used to establish an invariant space; some vertices of the fine level are selected as feature vertices to carry the watermarking information. The vertices at the coarse level are not used in watermark embedding; therefore, the embedded watermark does not affect the established coordinate space and normalization parameters are available to embed the watermark in feature vertices. These feature vertices located in the rapidly changing regions ensure that the watermark has better invisibility. The classification of the vertices and the choice of the feature vertices provide a suitable trade-off between good transparency and maximum robustness. The experimental results show that in comparison with other well-known 3D-model watermarking algorithms, our method can resist common attacks such as noise, smoothing, simplification, cropping, rotation, translation, and scaling while demonstrating good imperceptibility.