基于非局部相似性的三维模型去噪

Mesh denoising based on non-local similarity

随着计算机技术的发展,人们对于高质量三维模型的需求也日渐增加。然而,高质量的三维模型通常是通过扫描设备获取的,而扫描获取的模型不可避免的带有噪声,因此三维模型去噪技术已经成为计算机图形学中的热门研究方向。针对三维模型具有的非局部相似性,本文提出了一种基于k环结构的寻找相似结构的方法,以及一种基于非局部相似性的三维模型面法向量滤波算法。实验结果表明,该算法在主观与客观对比中均可以达到良好的效果,且在特征恢复方面有明显的优势。

With the development of computer science,the demand for high quality 3D models has been increased.However,high quality models are usually obtained by scanners,which are inevitably contaminated by noise.Hence,mesh denoising has become a hot research topic in the area of computer graphic.In this paper,aiming at the non-local similarity of mesh,a k-ring patch based scheme for finding similar structures and a non-local similarity based face normal filtering scheme are proposed.The experiment results show our scheme achieves satisfying results on both objective and subjective comparisons and has obvious advantage on recovering feature.