딥 러닝 기반 얼굴 메쉬 데이터 디노이징 시스템
A Deep Learning-Based Face Mesh Data Denoising System
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
3차원 프린터나 깊이 카메라 등을 이용하면 실세계의 3차원 메쉬 데이터를 손쉽게 생성할 수 있지만, 이렇게 생성된 데이터에는 필연적으로 불필요한 노이즈가 포함되어 있다. 따라서, 온전한 3차원 메쉬 데이터를 얻기 위해서는 메쉬 디노이징 작업이 필수적이다. 하지만 기존의 수학적인 디노이징 방법들은 전처리 작업이 필요하며 3차원 메쉬의 일부 중요한 특징들이사라지는 문제점이 있다. 본 논문에서는 이러한 문제를 해결하기 위해 딥 러닝 기반의 3차원 메쉬 디노이징 기법을 소개한다. 구체적으로 본 논문에서는 인코더와 디코더로 구성된 컨볼루션 기반 오토인코더 모델을 제안한다. 메쉬 데이터에 적용하는컨볼루션 연산은 메쉬 데이터를 구성하고 있는 각각의 정점과 그 주변의 정점들 간의 관계를 고려하여 디노이징을 수행하며,컨볼루션이 완료되면 학습 속도 향상을 위해 샘플링 연산을 수행한다. 실험 결과, 본 논문에서 제안한 오토인코더 모델이 기존 방식보다 더 빠르고 더 높은 품질의 디노이징된 데이터를 생성함을 확인하였다.
Abstract
Although one can easily generate real-world 3D mesh data using a 3D printer or a depth camera, the generated datainevitably includes unnecessary noise. Therefore, mesh denoising is essential to obtain intact 3D mesh data. However,conventional mathematical denoising methods require preprocessing and often eliminate some important features of the3D mesh. To address this problem, this paper proposes a deep learning based 3D mesh denoising method. Specifically,we propose a convolution-based autoencoder model consisting of an encoder and a decoder. The convolution operationapplied to the mesh data performs denoising considering the relationship between each vertex constituting the meshdata and the surrounding vertices. When the convolution is completed, a sampling operation is performed to improvethe learning speed. Experimental results show that the proposed autoencoder model produces faster and higher qualitydenoised data than the conventional methods.
참고문헌 (16)
- G. Taubin, "A signal processing approach to fair surface design", SIGGRAPH, pp.351~358 (1995)
- M. Desbrun, "Implicit fairing of irregular meshes using diffusion and curvature flow", SIGGRAPH, pp.317~324 (1999)
- A. Nealen, "Laplacian mesh optimization", Proceedings of the 4th international conference on Computer graphics and interactive techniques in Australasia and Southeast Asia (GRAPHITE ’06), pp.381~389 (2006)
- U. Clarenz, "Anisotropic geometric diffusion in surface processing", Proc. of the conference on Visualization, pp.397~405 (2000)
- T. Tasdizen, "Geometric surface smoothing via anisotropic diffusion of normals", Proc. of the conference on Visualization, pp.125~132 (2002)
- Chandrajit L. Bajaj, "Anisotropic diffusion of surfaces and functions on surfaces", ACM Transactions on Graphics, 22(1), pp.4~32 (2003), Association for Computing Machinery (ACM)[DOI]
- H. Yagou, "Mesh denoising via iterative alpha-trimming and nonlinear diffusion of normals with automatic thresholding", Computer Graphics International, pp.28~33 (2003)
- Klaus Hildebrandt, "Anisotropic Filtering of Non-Linear Surface Features", Computer Graphics Forum, 23(3), pp.391~400 (2004), Wiley[DOI]
- Thouis R. Jones, "Non-iterative, feature-preserving mesh smoothing", ACM Transactions on Graphics, 22(3), pp.943~949 (2003), Association for Computing Machinery (ACM)[DOI]
- Shachar Fleishman, "Bilateral mesh denoising", ACM Transactions on Graphics, 22(3), pp.950~953 (2003), Association for Computing Machinery (ACM)[DOI]
- H. Yagou, "Mesh smoothing via mean and median filtering applied to face normals", Geom. Model. and Proc., pp.124~131 (2002)
- Yuzhong Shen, "Fuzzy vector median-based surface smoothing", IEEE Transactions on Visualization and Computer Graphics, 10(3), pp.252~265 (2004), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Hanqi Fan, "Robust Feature-Preserving Mesh Denoising Based on Consistent Subneighborhoods", IEEE Transactions on Visualization and Computer Graphics, 16(2), pp.312~324 (2010), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Xuequan Lu, "A Robust Scheme for Feature-Preserving Mesh Denoising", IEEE Transactions on Visualization and Computer Graphics, 22(3), pp.1181~1194 (2016), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Mingqiang Wei, "Bi-Normal Filtering for Mesh Denoising", IEEE Transactions on Visualization and Computer Graphics, 21(1), pp.43~55 (2015), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- A. Ranjan, "Generating 3D Faces Using Convolutional Mesh Autoencoders", ECCV 2018, pp.725~741 (2018)