mlsanityData to 14 Oct 2026

MeshCNN Fundamentals: Geometric Learning through a Reconstructable Representation

arXiv 2105.13277 ↗First posted 27 May 2021cs.GR cs.CG cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Mesh-based learning is one of the popular approaches nowadays to learn shapes. The most established backbone in this field is MeshCNN. In this paper, we propose infusing MeshCNN with geometric reasoning to achieve higher quality learning. Through careful analysis of the way geometry is represented through-out the network, we submit that this representation should be rigid motion invariant, and should allow reconstructing the original geometry. Accordingly, we introduce the first and second fundamental forms as an edge-centric, rotation and translation invariant, reconstructable representation. In addition, we update the originally proposed pooling scheme to be more geometrically driven. We validate our analysis through experimentation, and present consistent improvement upon the MeshCNN baseline, as well as other more elaborate state-of-the-art architectures. Furthermore, we demonstrate this fundamental forms-based representation opens the door to accessible generative machine learning over meshes.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 39 of its 72 references that are in the map. First posted on arXiv 27 May 2021.

Builds on

39 of 72 references.

References · 39

  1. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,361
  2. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  3. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  4. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,689
  5. 20163D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotationcited by 1,476
  6. 2019KPConv: Flexible and Deformable Convolution for Point Cloudscited by 1,205
  7. 2016Geometric deep learning: going beyond Euclidean datacited by 1,048
  8. 2018PointCNN: Convolution On X-Transformed Pointscited by 764
  9. 2019Relation-Shape Convolutional Neural Network for Point Cloud Analysiscited by 346
  10. 2017Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphscited by 310
  11. 2015Deep Convolutional Networks on Graph-Structured Datacited by 246
  12. 2016Volumetric and Multi-View CNNs for Object Classification on 3D Datacited by 240
27 more
  1. 2017SEGCloud: Semantic Segmentation of 3D Point Cloudscited by 179
  2. 20193D Point Cloud Generative Adversarial Network Based on Tree Structured Graph Convolutionscited by 173
  3. 2019PU-GAN: a Point Cloud Upsampling Adversarial Networkcited by 164
  4. 2018Point Convolutional Neural Networks by Extension Operatorscited by 154
  5. 20203D-MPA: Multi Proposal Aggregation for 3D Semantic Instance Segmentationcited by 141
  6. 2016Generative and Discriminative Voxel Modeling with Convolutional Neural Networkscited by 128
  7. 2019DensePoint: Learning Densely Contextual Representation for Efficient Point Cloud Processingcited by 104
  8. 2017PCPNET: Learning Local Shape Properties from Raw Point Cloudscited by 98
  9. 2016Learning shape correspondence with anisotropic convolutional neural networkscited by 95
  10. 2019SpiralNet++: A Fast and Highly Efficient Mesh Convolution Operatorcited by 89
  11. 2018Triplet-Center Loss for Multi-View 3D Object Retrievalcited by 69
  12. 2020Neural Mesh Flow: 3D Manifold Mesh Generation via Diffeomorphic Flowscited by 61
  13. 2020Fully Convolutional Mesh Autoencoder using Efficient Spatially Varying Kernelscited by 53
  14. 2020Deep Geometric Texture Synthesiscited by 51
  15. 2020Neural Subdivisioncited by 49
  16. 2020Primal-Dual Mesh Convolutional Neural Networkscited by 44
  17. 2020DualConvMesh-Net: Joint Geodesic and Euclidean Convolutions on 3D Meshescited by 42
  18. 2018A Simple Approach to Intrinsic Correspondence Learning on Unstructured 3D Meshescited by 39
  19. 2018Multi-directional Geodesic Neural Networks via Equivariant Convolutioncited by 32
  20. 2016GIFT: A Real-time and Scalable 3D Shape Search Enginecited by 31
  21. 2019Dominant Set Clustering and Pooling for Multi-View 3D Object Recognitioncited by 25
  22. 2018Surface Networks via General Coverscited by 22
  23. 2016Orientation-boosted Voxel Nets for 3D Object Recognitioncited by 17
  24. 2019Mesh Variational Autoencoders with Edge Contraction Poolingcited by 14
  25. 2018Classification of Point Cloud Scenes with Multiscale Voxel Deep Networkcited by 11
  26. 2018Learning 3D Shapes as Multi-Layered Height-maps using 2D Convolutional Networkscited by 9
  27. 2018SPNet: Deep 3D Object Classification and Retrieval using Stereographic Projectioncited by 8

Built on by

Not cited yet.