Effective Rotation-Invariant Point CNN with Spherical Harmonics Kernels

@article{Poulenard2019EffectiveRP,
  title={Effective Rotation-Invariant Point CNN with Spherical Harmonics Kernels},
  author={Adrien Poulenard and Marie-Julie Rakotosaona and Yann Ponty and Maks Ovsjanikov},
  journal={2019 International Conference on 3D Vision (3DV)},
  year={2019},
  pages={47-56}
}
We present a novel rotation invariant architecture operating directly on point cloud data. [...] Key Method We achieve this by employing a spherical harmonics-based kernel at different layers of the network, which is guaranteed to be invariant to rigid motions. We also introduce a more efficient pooling operation for PCNN using space-partitioning data-structures. This results in a flexible, simple and efficient architecture that achieves accurate results on challenging shape analysis tasks, including…Expand
LFNet: Local Rotation Invariant Coordinate Frame for Robust Point Cloud Analysis
Deep neural networks have achieved great progress in 3D scene understanding. However, recent methods mainly focused on objects with canonical orientations in contrast with random postures in reality.Expand
Rotation-Invariant Point Convolution With Multiple Equivariant Alignments.
  • Hugues Thomas
  • Computer Science
  • 2020 International Conference on 3D Vision (3DV)
  • 2020
TLDR
It is shown that using rotation-equivariant alignments, it is possible to make any convolutional layer rotation-invariant, and this core layer is designed which improves state-of-the-art results in both object classification and semantic segmentation and reduces the gap between rotation- Invariant and standard 3D deep learning approaches. Expand
Designing Rotationally Invariant Neural Networks from PDEs and Variational Methods
TLDR
The goal of this paper is to investigate how diffusion and variational models achieve rotation invariance and transfer these ideas to neural networks, and to propose activation functions which couple network channels by combining information from several oriented filters. Expand
CNNs on surfaces using rotation-equivariant features
TLDR
A network architecture for surfaces that consists of vector-valued, rotation-equivariant features that makes it possible to locally align features, which were computed in arbitrary coordinate systems, when aggregating features in a convolution layer. Expand
You Only Hypothesize Once: Point Cloud Registration with Rotation-equivariant Descriptors
TLDR
A novel local descriptor-based framework, called You Only Hypothesize Once (YOHO), for the registration of two unaligned point clouds, which achieves the rotation invariance by recent technologies of group equivariant feature learning, which brings more robustness to point density and noise. Expand
Concentric Spherical GNN for 3D Representation Learning
TLDR
A novel multi-resolution convolutional architecture for learning over concentric spherical feature maps, of which the single sphere representation is a special case, and an efficient mapping of point clouds to concentrics spherical images using radial basis functions, thereby bridging spherical convolutions on grids with general point clouds. Expand
Diffusion is All You Need for Learning on Surfaces
TLDR
The principled, geometric nature of these networks makes them agnostic to the underlying representation and insensitive to discretization, which means significant robustness to mesh sampling, and even the ability to train on a mesh and evaluate on a point cloud. Expand
Robust Object Classification Approach using Spherical Harmonics
TLDR
The proposed spherical convolution neural network framework uses the voxel grid of concentric spheres to learn features over the unit ball that are less sensitive to data augmentation due to the selected sampling strategy and the designed convolution operation. Expand
Multi-Head Attentional Point Cloud Classification and Segmentation Using Strictly Rotation-Invariant Representations
TLDR
A novel neural network called Multi-head Attentional Point Cloud Classification and Segmentation Using Strictly Rotation-invariant Representations that realizes better point cloud classification and segmentation performance than most state-of-the-art methods. Expand
RotPredictor: Unsupervised Canonical Viewpoint Learning for Point Cloud Classification
TLDR
This work proposes to rotate the point cloud into a canonical viewpoint for boosting the following downstream target task, e.g., object classification and part segmentation, and is an independent plug and play module, which can be employed by any point-based deep learning framework without extra burden. Expand
...
1
2
3
4
...

References

SHOWING 1-10 OF 39 REFERENCES
Pointwise Rotation-Invariant Network with Adaptive Sampling and 3D Spherical Voxel Convolution
TLDR
A brand new point-set learning framework PRIN is proposed, namely, Pointwise Rotation-Invariant Network, focusing on rotation-invariant feature extraction in point clouds analysis, which shows that, on the dataset with randomly rotated point clouds, PRIN demonstrates better performance than state-of-the-art methods without any data augmentation. Expand
PRIN: Pointwise Rotation-Invariant Network
TLDR
This paper proposes a new point-set learning framework named Pointwise Rotation-Invariant Network (PRIN), focusing on achieving rotation-invariance in point clouds, and shows performance better than state-of-the-art methods on part segmentation without data augmentation. Expand
Exploring local rotation invariance in 3D CNNs with steerable filters
TLDR
The results show the importance of LRI in CNNs and the need for a fine rotation sampling and a drastic reduction of trainable parameters and of convolution operations, as well as avoiding approximations due to interpolation of rotated kernels. Expand
CubeNet: Equivariance to 3D Rotation and Translation
TLDR
A Group Convolutional Neural Network with linear equivariance to translations and right angle rotations in three dimensions is introduced, and is believed to be the first 3D rotation equivariant CNN for voxel representations. Expand
PointCNN: Convolution On X-Transformed Points
TLDR
This work proposes to learn an Χ-transformation from the input points to simultaneously promote two causes: the first is the weighting of the input features associated with the points, and the second is the permutation of the points into a latent and potentially canonical order. Expand
PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation
TLDR
This paper designs a novel type of neural network that directly consumes point clouds, which well respects the permutation invariance of points in the input and provides a unified architecture for applications ranging from object classification, part segmentation, to scene semantic parsing. Expand
PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space
TLDR
A hierarchical neural network that applies PointNet recursively on a nested partitioning of the input point set and proposes novel set learning layers to adaptively combine features from multiple scales to learn deep point set features efficiently and robustly. Expand
Spatial Transformer Networks
TLDR
This work introduces a new learnable module, the Spatial Transformer, which explicitly allows the spatial manipulation of data within the network, and can be inserted into existing convolutional architectures, giving neural networks the ability to actively spatially transform feature maps. Expand
Learning Steerable Filters for Rotation Equivariant CNNs
TLDR
Steerable Filter CNNs (SFCNNs) are developed which achieve joint equivariance under translations and rotations by design and generalize He's weight initialization scheme to filters which are defined as a linear combination of a system of atomic filters. Expand
Geodesic Convolutional Neural Networks on Riemannian Manifolds
TLDR
Geodesic Convolutional Neural Networks (GCNN), a generalization of the convolutional neural networks (CNN) paradigm to non-Euclidean manifolds is introduced, allowing to achieve state-of-the-art performance in problems such as shape description, retrieval, and correspondence. Expand
...
1
2
3
4
...