mlsanityData to 7 Oct 2026

FA-KPConv: Introducing Euclidean Symmetries to KPConv via Frame Averaging

arXiv 2505.04485 ↗First posted 7 May 2025cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

We present Frame-Averaging Kernel-Point Convolution (FA-KPConv), a neural network architecture built on top of the well-known KPConv, a widely adopted backbone for 3D point cloud analysis. Even though invariance and/or equivariance to Euclidean transformations are required for many common tasks, KPConv-based networks can only approximately achieve such properties when training on large datasets or with significant data augmentations. Using Frame Averaging, we allow to flexibly customize point cloud neural networks built with KPConv layers, by making them exactly invariant and/or equivariant to translations, rotations and/or reflections of the input point clouds. By simply wrapping around an existing KPConv-based network, FA-KPConv embeds geometrical prior knowledge into it while preserving the number of learnable parameters and not compromising any input information. We showcase the benefit of such an introduced bias for point cloud classification and point cloud registration, especially in challenging cases such as scarce training data or randomly rotated test data.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 39 of its 45 references that are in the map. First posted on arXiv 7 May 2025.

Builds on

39 of 45 references.

References · 39

  1. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  2. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  3. 2018Dynamic Graph CNN for Learning on Point Cloudscited by 2,220
  4. 2020Point Transformercited by 1,462
  5. 2015Spatial Transformer Networkscited by 1,423
  6. 2019KPConv: Flexible and Deformable Convolution for Point Cloudscited by 1,202
  7. 2016Group Equivariant Convolutional Networkscited by 1,154
  8. 2021E(n) Equivariant Graph Neural Networkscited by 951
  9. 2018PointCNN: Convolution On X-Transformed Pointscited by 764
  10. 2018Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point cloudscited by 735
  11. 2020PCT: Point cloud transformercited by 730
  12. 2018PointConv: Deep Convolutional Networks on 3D Point Cloudscited by 656
27 more
  1. 2021Point-BERT: Pre-training 3D Point Cloud Transformers with Masked Point Modelingcited by 647
  2. 2020SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networkscited by 602
  3. 2023Point Transformer V3: Simpler, Faster, Strongercited by 579
  4. 2022Masked Autoencoders for Point Cloud Self-supervised Learningcited by 456
  5. 2018Spherical CNNscited by 452
  6. 2022PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategiescited by 432
  7. 2022Point Transformer V2: Grouped Vector Attention and Partition-based Poolingcited by 384
  8. 20163DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructionscited by 333
  9. 2017Dynamic Edge-Conditioned Filters in Convolutional Neural Networks on Graphscited by 307
  10. 2021Vector Neurons: A General Framework for SO(3)-Equivariant Networkscited by 304
  11. 2016Steerable CNNscited by 298
  12. 2022Point-M2AE: Multi-scale Masked Autoencoders for Hierarchical Point Cloud Pre-trainingcited by 263
  13. 2018SpiderCNN: Deep Learning on Point Sets with Parameterized Convolutional Filterscited by 261
  14. 2021PAConv: Position Adaptive Convolution with Dynamic Kernel Assembling on Point Cloudscited by 211
  15. 2018PPFNet: Global Context Aware Local Features for Robust 3D Point Matchingcited by 180
  16. 2021Frame Averaging for Invariant and Equivariant Network Designcited by 166
  17. 2022REGTR: End-to-end Point Cloud Correspondences with Transformerscited by 94
  18. 2023GeoTransformer: Fast and Robust Point Cloud Registration with Geometric Transformercited by 83
  19. 2021Equivariant Point Network for 3D Point Cloud Analysiscited by 81
  20. 2019Rotation Invariant Convolutions for 3D Point Clouds Deep Learningcited by 80
  21. 2021Adaptive Graph Convolution for Point Cloud Analysiscited by 60
  22. 2022GFNet: Geometric Flow Network for 3D Point Cloud Semantic Segmentationcited by 36
  23. 2019SRINet: Learning Strictly Rotation-Invariant Representations for Point Cloud Classification and Segmentationcited by 29
  24. 2023Advancements in Point Cloud Data Augmentation for Deep Learning: A Surveycited by 24
  25. 2020Boosting Deep Neural Networks with Geometrical Prior Knowledge: A Surveycited by 12
  26. 2022ObPose: Leveraging Pose for Object-Centric Scene Inference and Generation in 3Dcited by 2
  27. 2023A Strong Baseline for Point Cloud Registration via Direct Superpoints Matchingcited by 1

Built on by

Not cited yet.