Deep Mesh Prior: Unsupervised Mesh Restoration using Graph Convolutional Networks
Not on the map. Why?
Abstract (arXiv)
This paper addresses mesh restoration problems, i.e., denoising and completion, by learning self-similarity in an unsupervised manner. For this purpose, the proposed method, which we refer to as Deep Mesh Prior, uses a graph convolutional network on meshes to learn the self-similarity. The network takes a single incomplete mesh as input data and directly outputs the reconstructed mesh without being trained using large-scale datasets. Our method does not use any intermediate representations such as an implicit field because the whole process works on a mesh. We demonstrate that our unsupervised method performs equally well or even better than the state-of-the-art methods using large-scale datasets.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 35 of its 55 references that are in the map. First posted on arXiv 2 Jul 2021.
Builds on
35 of 55 references.
References · 35
- 2014Adam: A Method for Stochastic Optimizationcited by 37,240
- 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
- 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
- 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
- 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
- 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,689
- 2019Fast Graph Representation Learning with PyTorch Geometriccited by 2,434
- 2016Convolutional Neural Networks on Graphs with Fast Localized Spectral Filteringcited by 2,089
- 2015Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networkscited by 2,054
- 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,041
- 2017Deep Image Priorcited by 1,098
- 2018Learning Implicit Fields for Generative Shape Modelingcited by 1,060
23 more
- 2016Geometric deep learning on graphs and manifolds using mixture model CNNscited by 468
- 2018PCN: Point Completion Networkcited by 384
- 2016Shape Completion using 3D-Encoder-Predictor CNNs and Shape Synthesiscited by 219
- 2018PU-Net: Point Cloud Upsampling Networkcited by 209
- 2019BSP-Net: Generating Compact Meshes via Binary Space Partitioningcited by 206
- 2017Octree Generating Networks: Efficient Convolutional Architectures for High-resolution 3D Outputscited by 189
- 2019CvxNet: Learnable Convex Decompositioncited by 174
- 2018Deep Geometric Prior for Surface Reconstructioncited by 110
- 2020Point Cloud Completion by Skip-attention Network with Hierarchical Foldingcited by 106
- 2018Patch-based Progressive 3D Point Set Upsamplingcited by 101
- 2019PointCleanNet: Learning to Denoise and Remove Outliers from Dense Point Cloudscited by 99
- 2017PCPNET: Learning Local Shape Properties from Raw Point Cloudscited by 98
- 2019BAE-NET: Branched Autoencoder for Shape Co-Segmentationcited by 89
- 2016Shape Completion Enabled Robotic Graspingcited by 79
- 2018Learning Shape Priors for Single-View 3D Completion and Reconstructioncited by 68
- 2019Unpaired Point Cloud Completion on Real Scans using Adversarial Trainingcited by 66
- 2018Scan2Mesh: From Unstructured Range Scans to 3D Meshescited by 51
- 2020Point2Mesh: A Self-Prior for Deformable Meshescited by 41
- 2020Deep Octree-based CNNs with Output-Guided Skip Connections for 3D Shape and Scene Completioncited by 33
- 2018Adaptive O-CNN: A Patch-based Deep Representation of 3D Shapescited by 25
- 2017Shape Inpainting using 3D Generative Adversarial Network and Recurrent Convolutional Networkscited by 23
- 20173D Reconstruction of Incomplete Archaeological Objects Using a Generative Adversarial Networkcited by 12
- 2020DNF-Net: a Deep Normal Filtering Network for Mesh Denoisingcited by 10
Built on by
Citing papers per year: 2023: 1; 2024: 1.
2 papers.