Feature-preserving mesh simplification through anisotropic Nyquist-based adaptive sampling of points inside the segmented regions

A wide increase of 3D applications for using mobile phones and other electrical devices reveals the importance of 3D mesh representation. Since visualization and implementation of a coarse and simplified mesh are easier than analyzing a high-resolution mesh, the simplified mesh is preferred for processing. In this paper, a new 3D mesh simplification method is presented to simplify a mesh by anisotropic Nyquist-based adaptive sampling of each segmented region on the surface. Since the sampling step is completed for each segmented region individually, the algorithm can preserve the sharp features of each segment, precisely. The least number of samples is selected from each segment based on its details. Adjusting the sampling procedure according to the geometrical features of the mesh leads to accurately approximate the overall shape of the original model. In order to connect the selected samples, the original mesh connections are employed to better maintain the structure and shape of the input mesh. The improved quality of the results obtained by the proposed method demonstrates its ability in better preserving fine-scale features of different complex meshes in comparison with the previous studies. The simplified models can be efficiently reconstructed based on the selected samples of each region.