Feature-preserving non-local denoising of static and time-varying range data

Abstract

We present a new method for noise removal on static and timevarying range data. Our approach predicts the restored position of a perturbed vertex using similar vertices in its neighborhood. It defines the required similarity measure in a new non-local fashion which compares regions of the surface instead of point pairs. This allows our algorithm to obtain a more accurate denoising result than previous state-of-the-art approaches and, at the same time, to better preserve fine features of the surface. Furthermore, our approach is easy to implement, effective, and flexibly applicable to different types of scanned data. We demonstrate this on several static and interesting new time-varying datasets obtained using laser and structured light scanners.

Fusion of range and color images for denoising and resolution enhancement with a non-local filter

“…The NL-Means filter was also used for simultaneous denoising and depth reconstruction from noisy image pairs [16]. Schall et al [17] were the first to apply NL-Means for denoising of 3D point data. They suppose that the data is given on a (not necessarily regular) 2D grid and compute the similarity weight according to the Euclidian distance in 3D.…”

“…In general, the weighting scheme leads to an effect similar to the adoption of NLMeans to 3D point data. The authors of [17] consider Euclidian distances in 3D instead of a Gaussian weighting in pixel coordinates (Eq. (3)) and therefore minimize the influence of pixels that belong to very distant regions in depth.…”

Noise analysis and synthesis for 3D laser depth scanners

“…To remove such noise from surface mesh models, many mesh denoising algorithms have been developed-see [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16][17][18][19] and other references therein. In evaluating the effectiveness of such denoising algorithms, both visual and numerical comparisons are used [12].…”

Correção de normais para suavização de nuvens de pontos

“…Outras técnicas para suavizar nuvens de pontos, baseadas na média não-local (Buades et al, 2005), foram propostas por Guillemot et al (2012);Schall et al (2007). Schall et al (2007) propõem um filtro bilateral, onde um dos pesos penaliza a distância dos pontos e o outro peso, chamado peso de similaridade, compara a semelhança entre as vizinhanças geométricas.…”

“…Schall et al (2007) propõem um filtro bilateral, onde um dos pesos penaliza a distância dos pontos e o outro peso, chamado peso de similaridade, compara a semelhança entre as vizinhanças geométricas.…”