mlsanityData to 14 Oct 2026

Level-Set Parameters: Novel Representation for 3D Shape Analysis

arXiv 2412.13502 ↗First posted 18 Dec 2024cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

3D shape analysis has been largely focused on traditional 3D representations of point clouds and meshes, but the discrete nature of these data makes the analysis susceptible to variations in input resolutions. Recent development of neural fields brings in level-set parameters from signed distance functions as a novel, continuous, and numerical representation of 3D shapes, where the shape surfaces are defined as zero-level-sets of those functions. This motivates us to extend shape analysis from the traditional 3D data to these novel parameter data. Since the level-set parameters are not Euclidean like point clouds, we establish correlations across different shapes by formulating them as a pseudo-normal distribution, and learn the distribution prior from the respective dataset. To further explore the level-set parameters with shape transformations, we propose to condition a subset of these parameters on rotations and translations, and generate them with a hypernetwork. This simplifies the pose-related shape analysis compared to using traditional data. We demonstrate the promise of the novel representations through applications in shape classification (arbitrary poses), retrieval, and 6D object pose estimation.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 62 of its 82 references that are in the map. First posted on arXiv 18 Dec 2024.

Builds on

62 of 82 references.

References · 62

  1. 2015Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shiftcited by 7,005
  2. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  3. 2017Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networkscited by 4,897
  4. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,401
  5. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,897
  6. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,689
  7. 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,238
  8. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,223
  9. 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,041
  10. 2021NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstructioncited by 1,712
  11. 2020Point Transformercited by 1,470
  12. 2020NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collectionscited by 1,232
38 more
  1. 2019KPConv: Flexible and Deformable Convolution for Point Cloudscited by 1,205
  2. 2018Learning Implicit Fields for Generative Shape Modelingcited by 1,060
  3. 2019An Introduction to Variational Autoencoderscited by 1,000
  4. 2021Volume Rendering of Neural Implicit Surfacescited by 996
  5. 2018PointCNN: Convolution On X-Transformed Pointscited by 764
  6. 2020Convolutional Occupancy Networkscited by 761
  7. 20173D Semantic Segmentation with Submanifold Sparse Convolutional Networkscited by 748
  8. 2018Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point cloudscited by 746
  9. 2020PCT: Point cloud transformercited by 732
  10. 2020Implicit Geometric Regularization for Learning Shapescited by 710
  11. 2020pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesiscited by 708
  12. 2018PointConv: Deep Convolutional Networks on 3D Point Cloudscited by 657
  13. 2017Deep Neural Networks as Gaussian Processescited by 622
  14. 2021Neural Fields in Visual Computing and Beyondcited by 540
  15. 2019Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimationcited by 495
  16. 2022MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstructioncited by 470
  17. 2018Spherical CNNscited by 456
  18. 2019SAL: Sign Agnostic Learning of Shapes from Raw Datacited by 390
  19. 2022Point Transformer V2: Grouped Vector Attention and Partition-based Poolingcited by 388
  20. 2018PVNet: Pixel-wise Voting Network for 6DoF Pose Estimationcited by 385
  21. 2019DenseFusion: 6D Object Pose Estimation by Iterative Dense Fusioncited by 355
  22. 2020Implicit Functions in Feature Space for 3D Shape Reconstruction and Completioncited by 343
  23. 2016OctNet: Learning Deep 3D Representations at High Resolutionscited by 342
  24. 20163DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructionscited by 336
  25. 2019Deep Closest Point: Learning Representations for Point Cloud Registrationcited by 321
  26. 2021Vector Neurons: A General Framework for SO(3)-Equivariant Networkscited by 305
  27. 2020PREDATOR: Registration of 3D Point Clouds with Low Overlapcited by 290
  28. 2020TEASER: Fast and Certifiable Point Cloud Registrationcited by 274
  29. 2017Learning SO(3) Equivariant Representations with Spherical CNNscited by 248
  30. 2020Learned Initializations for Optimizing Coordinate-Based Neural Representationscited by 242
  31. 2018Deep Object Pose Estimation for Semantic Robotic Grasping of Household Objectscited by 241
  32. 2019PointNetLK: Robust & Efficient Point Cloud Registration using PointNetcited by 233
  33. 2019Pix2Pose: Pixel-Wise Coordinate Regression of Objects for 6D Pose Estimationcited by 218
  34. 2020MetaSDF: Meta-learning Signed Distance Functionscited by 214
  35. 2022From data to functa: Your data point is a function and you can treat it like onecited by 187
  36. 2016Go-ICP: A Globally Optimal Solution to 3D ICP Point-Set Registrationcited by 159
  37. 2023HyperDiffusion: Generating Implicit Neural Fields with Weight-Space Diffusioncited by 152
  38. 2017Escape from Cells: Deep Kd-Networks for the Recognition of 3D Point Cloud Modelscited by 142

50 most cited of 62 All 62 (JSON)

Built on by

Not cited yet.