Data-driven Geometry-recovering Mesh Denoising. (September 2019)

• Record Type: Journal Article
• Authors: Wang, Jun; Huang, Jin; Wang, Fu Lee; Wei, Mingqiang; Xie, Haoran; Qin, Jing
• Is Part Of: Computer aided design. Volume 114(2019)
• Page Start: 133
• Page End: 142
• Publication Date: 2019-09
• Subjects: Mesh denoising -- Geometry-recovering -- Low-rank matrix recovery -- Reverse filter -- Normal variation learning; Computer-aided design -- Periodicals; Engineering design -- Data processing -- Periodicals; Computer graphics -- Periodicals; Conception technique -- Informatique -- Périodiques; Infographie -- Périodiques; Computer graphics; Engineering design -- Data processing; Periodicals; Electronic journals; 620.00420285
• DOI: 10.1016/j.cad.2019.05.027
• Languages: English
• ISSNs: 0010-4485

Abstract

Abstract: Depth cameras and 3D scanners significantly simplify the procedure of geometric modeling. 3D surfaces have become more widespread, leading to a great demand for noise removal with the expectation of the minimal disturbance of mesh geometry. We propose a novel two-step data-driven mesh denoising approach. The first step removes noise by learning normal variations from noisy models to their ground-truth counterparts. Unlike existing denoising methods, we present the second step to recover the mesh geometry lost in the first step. The second step understands the commonly used filters by learning the mapping from filtered models to their ground-truth counterparts. In addition, (1) to handle noise with large variations, we model normal estimation as a low-rank matrix recovery problem in similar-patch collaboration before the first-step learning; (2) to recover the real geometry of a denoised mesh, we reversely filter the denoised mesh to obtain more geometry cues before the second-step learning. The detailed quantitative and qualitative results on various data demonstrate that, our two-step learning algorithm competes favorably with the state-of-the-art methods in terms of mesh geometry preservation and noise-robustness. Highlights: We design a two-step normal variation learning method for geometry-recovering mesh denoising. We collect similar patches and formulate them as a low-rank matrix recovery problem to estimate surface normals for dealing with different levels of noise. We present a reverse normal filter for finding more geometry cues from filtered meshes.