Two-stage Point-sampled Model Denoising by Robust Ellipsoid Criterion and Mean Shift

Surfaces are often reconstructed from unorganized point sets with noise, so denoising is an essential step in creating perfect point-sampled models. A two-stage point clouds denoising method is presented which combines the ellipsoid criterion with the mean shift filtering approach. so denoising method is efficient for point-sampled model. Firstly, we determine if one point is the noise or not by the ellipsoid criterion. After acquiring new point sets being less noisy, we smooth the remains noise by mean shift point clouds denoising method. Experiments show that our method can smooth the noise efficiently.