Generated realistic noise and rotation-equivariant models for data-driven mesh denoising

DOI10.1016/J.CAGD.2024.102306MaRDI QIDQ6563921zbMATH OpenFDO

Authors Sipeng Yang, Wenhui Ren, Xiwen Zeng, Qingchuan Zhu, Hongbo Fu, Kaijun Fan, Lei Yang, Jingping Yu, Qilong Kou, Xiao-Gang Jin

Publication date 28 June 2024

Published in Computer Aided Geometric Design (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/J.CAGD.2024.102306

zbMATH Keywords

equivariant modelsmesh denoisingrealistic noise generation

Mathematics Subject Classification ID

Numerical approximation and computational geometry (primarily algorithms) (65Dxx)

Recommendations

• An attention enhanced dual graph neural network for mesh denoising
• A novel anisotropic second order regularization for mesh denoising
• 1D CNNs and face-based random walks: a powerful combination to enhance mesh understanding and 3D semantic segmentation
• TPNet: a novel mesh analysis method via topology preservation and perception enhancement

Cites work

• Beyond a Gaussian Denoiser: Residual Learning of Deep CNN for Image Denoising
• Computing geodesic paths on manifolds
• Mesh denoising via L ₀ minimization
• Robust mesh denoising via vertex pre-filtering and \(L_1\)-median normal filtering
• Rolling normal filtering for point clouds

Cited in

(6)

• An attention enhanced dual graph neural network for mesh denoising
• Point-styleGAN: multi-scale point cloud synthesis with style modulation
• FuncScene: function-centric indoor scene synthesis via a variational autoencoder framework
• Real-time volume rendering with octree-based implicit surface representation
• 1D CNNs and face-based random walks: a powerful combination to enhance mesh understanding and 3D semantic segmentation
• Robust and feature-preserving point denoising via adaptive scale selection mechanism

This page was built for publication: Generated realistic noise and rotation-equivariant models for data-driven mesh denoising

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6563921)