mlsanityData to 7 Oct 2026

Subdivision-Based Mesh Convolution Networks

arXiv 2106.02285 ↗First posted 4 Jun 2021cs.CV cs.GR cs.LGcited by 60

Not on the map. Why?

Abstract (arXiv)

Convolutional neural networks (CNNs) have made great breakthroughs in 2D computer vision. However, their irregular structure makes it hard to harness the potential of CNNs directly on meshes. A subdivision surface provides a hierarchical multi-resolution structure, in which each face in a closed 2-manifold triangle mesh is exactly adjacent to three faces. Motivated by these two observations, this paper presents SubdivNet, an innovative and versatile CNN framework for 3D triangle meshes with Loop subdivision sequence connectivity. Making an analogy between mesh faces and pixels in a 2D image allows us to present a mesh convolution operator to aggregate local features from nearby faces. By exploiting face neighborhoods, this convolution can support standard 2D convolutional network concepts, e.g. variable kernel size, stride, and dilation. Based on the multi-resolution hierarchy, we make use of pooling layers which uniformly merge four faces into one and an upsampling method which splits one face into four. Thereby, many popular 2D CNN architectures can be easily adapted to process 3D meshes. Meshes with arbitrary connectivity can be remeshed to have Loop subdivision sequence connectivity via self-parameterization, making SubdivNet a general approach. Extensive evaluation and various applications demonstrate SubdivNet's effectiveness and efficiency.

Key facts. Cited by 60 papers in the map (data through 7 Oct 2026). Builds on 49 of its 95 references that are in the map. First posted on arXiv 4 Jun 2021.

Builds on

49 of 95 references.

References · 49

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  3. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  4. 2017Graph Attention Networkscited by 8,854
  5. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  6. 2016The Cityscapes Dataset for Semantic Urban Scene Understandingcited by 4,427
  7. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  8. 2018Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentationcited by 3,210
  9. 2016Xception: Deep Learning with Depthwise Separable Convolutionscited by 2,320
  10. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  11. 2016Geometric deep learning: going beyond Euclidean datacited by 1,042
  12. 2016A Point Set Generation Network for 3D Object Reconstruction from a Single Imagecited by 972
37 more
  1. 2018PointCNN: Convolution On X-Transformed Pointscited by 764
  2. 2020PCT: Point cloud transformercited by 730
  3. 2018Pixel2Mesh: Generating 3D Mesh Models from Single RGB Imagescited by 649
  4. 2015Multi-view Convolutional Neural Networks for 3D Shape Recognitioncited by 564
  5. 2016Geometric deep learning on graphs and manifolds using mixture model CNNscited by 468
  6. 2018Generating 3D faces using Convolutional Mesh Autoencoderscited by 267
  7. 2018Tangent Convolutions for Dense Prediction in 3Dcited by 173
  8. 2020DiffusionNet: Discretization Agnostic Learning on Surfacescited by 163
  9. 20183D-CODED : 3D Correspondences by Deep Deformationcited by 160
  10. 2017Escape from Cells: Deep Kd-Networks for the Recognition of 3D Point Cloud Modelscited by 142
  11. 2017Deep Functional Maps: Structured Prediction for Dense Shape Correspondencecited by 134
  12. 2017VNect: Real-time 3D Human Pose Estimation with a Single RGB Cameracited by 129
  13. 2020Deep Geometric Functional Maps: Robust Feature Learning for Shape Correspondencecited by 125
  14. 2016Learning shape correspondence with anisotropic convolutional neural networkscited by 95
  15. 2018MeshNet: Mesh Neural Network for 3D Shape Representationcited by 95
  16. 2019SpiralNet++: A Fast and Highly Efficient Mesh Convolution Operatorcited by 89
  17. 2018Unsupervised Deep Learning for Structured Shape Matchingcited by 81
  18. 2018Continuous and Orientation-preserving Correspondences via Functional Mapscited by 77
  19. 2016SyncSpecCNN: Synchronized Spectral CNN for 3D Shape Segmentationcited by 70
  20. 2018Robust Watertight Manifold Surface Generation Method for ShapeNet Modelscited by 58
  21. 2018Scan2Mesh: From Unstructured Range Scans to 3D Meshescited by 51
  22. 2020Deep Geometric Texture Synthesiscited by 51
  23. 2020Neural Subdivisioncited by 49
  24. 2017O-CNN: Octree-based Convolutional Neural Networks for 3D Shape Analysiscited by 47
  25. 2018TextureNet: Consistent Local Parametrizations for Learning from High-Resolution Signals on Meshescited by 45
  26. 2020Primal-Dual Mesh Convolutional Neural Networkscited by 44
  27. 2020DualConvMesh-Net: Joint Geodesic and Euclidean Convolutions on 3D Meshescited by 42
  28. 2018A Simple Approach to Intrinsic Correspondence Learning on Unstructured 3D Meshescited by 39
  29. 2018Multi-directional Geodesic Neural Networks via Equivariant Convolutioncited by 32
  30. 2020MeshWalker: Deep Mesh Understanding by Random Walkscited by 32
  31. 2019ZoomOut: Spectral Upsampling for Efficient Shape Correspondencecited by 30
  32. 2018Surface Networks via General Coverscited by 22
  33. 2020A Survey on Deep Geometry Learning: From a Representation Perspectivecited by 19
  34. 2018Dual-Primal Graph Convolutional Networkscited by 18
  35. 2017Surface Networkscited by 14
  36. 2021HodgeNet: Learning Spectral Geometry on Triangle Meshescited by 13
  37. 2020DNF-Net: a Deep Normal Filtering Network for Mesh Denoisingcited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 2; 2022: 14; 2023: 14; 2024: 12; 2025: 10; 2026: 8.

60 papers.

Sort:

By area:

  1. 2023MeshGPT: Generating Triangle Meshes with Decoder-Only Transformerscited by 254
  2. 2023Deep Learning on Implicit Neural Representations of Shapescited by 62
  3. 2022A Survey on Graph Neural Networks and Graph Transformers in Computer Vision: A Task-Oriented Perspectivecited by 56
  4. 2024SCube: Instant Large-Scale Scene Reconstruction using VoxSplatscited by 56
  5. 20223D Highlighter: Localizing Regions on 3D Shapes via Text Descriptionscited by 32
  6. 2022MeshMAE: Masked Autoencoders for 3D Mesh Data Analysiscited by 31
  7. 2023Flexible Techniques for Differentiable Rendering with 3D Gaussianscited by 28
  8. 2024iSeg: Interactive 3D Segmentation via Interactive Attentioncited by 14
  9. 2022AGConv: Adaptive Graph Convolution on 3D Point Cloudscited by 13
  10. 2023Neural Progressive Meshescited by 12
  11. 2024SphereUFormer: A U-Shaped Transformer for Spherical 360 Perceptioncited by 12
  12. 2022Laplacian2Mesh: Laplacian-Based Mesh Understandingcited by 11
38 more
  1. 2023Single Mesh Diffusion Models with Field Latents for Texture Generationcited by 11
  2. 2024E(3)-Equivariant Mesh Neural Networkscited by 9
  3. 2021Mesh Convolutional Autoencoder for Semi-Regular Meshes of Different Sizescited by 8
  4. 2022TetGAN: A Convolutional Neural Network for Tetrahedral Mesh Generationcited by 8
  5. 2023Learning the Geodesic Embedding with Graph Neural Networkscited by 8
  6. 2023Neural Processing of Tri-Plane Hybrid Neural Fieldscited by 8
  7. 2025Bringing Attention to CAD: Boundary Representation Learning via Transformercited by 8
  8. 2022Keypoint Cascade Voting for Point Cloud Based 6DoF Pose Estimationcited by 7
  9. 2022Lattice Convolutional Networks for Learning Ground States of Quantum Many-Body Systemscited by 6
  10. 2022SphereDepth: Panorama Depth Estimation from Spherical Domaincited by 6
  11. 2022Neural Shape Compiler: A Unified Framework for Transforming between Text, Point Cloud, and Programcited by 6
  12. 2024Neural Laplacian Operator for 3D Point Cloudscited by 6
  13. 2023Deep Learning on Object-centric 3D Neural Fieldscited by 5
  14. 2021Mesh Convolution with Continuous Filters for 3D Surface Parsingcited by 4
  15. 2022Be Careful with Rotation: A Uniform Backdoor Pattern for 3D Shapecited by 4
  16. 2024TrIND: Representing Anatomical Trees by Denoising Diffusion of Implicit Neural Fieldscited by 4
  17. 2022Knee Cartilage Defect Assessment by Graph Representation and Surface Convolutioncited by 3
  18. 2022Transfer Learning using Spectral Convolutional Autoencoders on Semi-Regular Surface Meshescited by 3
  19. 2023GeoScaler: Geometry and Rendering-Aware Downsampling of 3D Mesh Texturescited by 3
  20. 2025ToothForge: Automatic Dental Shape Generation using Synchronized Spectral Embeddingscited by 3
  21. 2025SphereFusion: Efficient Panorama Depth Estimation via Gated Fusioncited by 2
  22. 2025Leveraging Point Transformers for Detecting Anatomical Landmarks in Digital Dentistrycited by 2
  23. 2025HodgeFormer: Transformers for Learnable Operators on Triangular Meshes through Data-Driven Hodge Matricescited by 2
  24. 2023NeuroExplainer: Fine-Grained Attention Decoding to Uncover Cortical Development Patterns of Preterm Infantscited by 1
  25. 2023SieveNet: Selecting Point-Based Features for Mesh Networkscited by 1
  26. 2023WrappingNet: Mesh Autoencoder via Deep Sphere Deformationcited by 1
  27. 2024An Intrinsic Vector Heat Networkcited by 1
  28. 2024Towards Secure and Usable 3D Assets: A Novel Framework for Automatic Visible Watermarkingcited by 1
  29. 2024A Recipe for Geometry-Aware 3D Mesh Transformerscited by 1
  30. 2025CageNet: A Meta-Framework for Learning on Wild Meshescited by 1
  31. 2025SSA3D: Text-Conditioned Assisted Self-Supervised Framework for Automatic Dental Abutment Designcited by 1
  32. 2026TIGaussian: Disentangle Gaussians for Spatial-Awared Text-Image-3D Alignmentcited by 1
  33. 2026SubdivAR: Autoregressive Next-Scale Prediction for Neural Mesh Subdivisioncited by 1
  34. 2022Brain Cortical Functional Gradients Predict Cortical Folding Patterns via Attention Mesh Convolutioncited by 0
  35. 2023Long Range Pooling for 3D Large-Scale Scene Understandingcited by 0
  36. 2023A Task-driven Network for Mesh Classification and Semantic Part Segmentationcited by 0
  37. 2024Self-Parameterization Based Multi-Resolution Mesh Convolution Networkscited by 0
  38. 2024ReFu: Recursive Fusion for Exemplar-Free 3D Class-Incremental Learningcited by 0

50 most cited of 60 All 60 (JSON)