Surface Reconstruction from High-density Points Using Deformed Grids

Fujimoto,K., Moriya,T., Nakayama,Y.

We present a method of surface reconstruction from unorganized high density points without a normal vector. The method needs a set of points, P = {p1, p2, ..., pN}, sampled on a surface in R^3. This method first sets a uniform grid and deforms each cell of the grid by fitting and moving the vertex of each cell to the nearest of the input points. Each cell has eight vertices, and because these vertices have two states whether they are moved or not, each cell has 28 = 256 patterns. After rotation is considered, the number of patterns becomes 14. Finally, It then constructs triangles according to the pattern of the vertices's state in each cell. Our method can work fast with little memory, and it is simple and easy to implement. We show that our method generated several polygon meshes from real-world range data.