In this paper, we propose a fast and efficient explicit three-dimensional (3D) mesh denoising algorithm that utilizes the Allen–Cahn (AC) equation with a fidelity term. The phase-field model is used to describe the characteristics of both the surface and interior of an object, allowing us to represent the 3D mesh model with noise using a phase-field function. By using the phase separation property of the AC equation and the fidelity term, the model can effectively preserve the original structures and features during the smoothing process, even in the presence of noise in various regions. The modified AC equation is numerically discretized using the explicit finite difference method, where the values at neighboring grid points are used as Dirichlet boundary conditions. Because the algorithm is local and explicit, it guarantees both effective denoising of 3D mesh models and rapid implementation speed. To validate the efficacy of the proposed algorithm, we conduct various computational experiments. Furthermore, we propose an implicit-explicit numerical scheme using the Crank–Nicolson method to address the denoising problem of 3D mesh models and perform related experiments.
Bibliographical noteFunding Information:
The first author Jian Wang expresses thanks for the Natural Science Foundation of the Jiangsu Higher Education Institutions of China (Grant no. 22KJB110020). The corresponding author (J.S. Kim) was supported by Korea University Grant. The authors greatly appreciate the reviewers for their constructive comments and suggestions, which have improved the quality of this paper.
© 2023 Elsevier Inc.
- 3D mesh denoising
- Allen–Cahn equation
ASJC Scopus subject areas
- Computational Mathematics
- Applied Mathematics