Rotation-Invariant Point Convolution With Multiple Equivariant Alignments
Recent attempts at introducing rotation invariance or equivariance in 3D deep learning approaches have shown promising results, but these methods still struggle to reach the performances of standard 3D neural networks. In this work we study the relation between equivariance and invariance in 3D point convolutions. We show that using rotation-equivariant alignments, it is possible to make any convolutional layer rotation-invariant. Furthermore, we improve this simple alignment procedure by using the alignment themselves as features in the convolution, and by combining multiple alignments together. With this core layer, we design rotation-invariant architectures which improve 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.
Code (0)
등록된 구현이 없습니다.
Tasks
Deep LearningSemantic SegmentationSimilar Papers 제목 키워드 기반
Rotation-Equivariant Keypoint Detection
We show how to train a rotation-equivariant representation to extract local keypoints for image matching. Existing learning-based methods focused on extracting translation-equivariant keypoints using conventional convolu…
Keypoint DetectionTranslationA Functional Approach to Rotation Equivariant Non-Linearities for Tensor Field Networks.
Learning pose invariant representation is a fundamental problem in shape analysis. Most existing deep learning algorithms for 3D shape analysis are not robust to rotations and are often trained on synthetic datasets …
PARE-Net: Position-Aware Rotation-Equivariant Networks for Robust Point Cloud Registration
Learning rotation-invariant distinctive features is a fundamental requirement for point cloud registration. Existing methods often use rotation-sensitive networks to extract features, while employing rotation augmentatio…
Inductive BiasPoint Cloud RegistrationPositionRelevance of Rotationally Equivariant Convolutions for Predicting Molecular Properties
Equivariant neural networks (ENNs) are graph neural networks embedded in $\mathbb{R}^3$ and are well suited for predicting molecular properties. The ENN library e3nn has customizable convolutions, which can be designed t…
Molecular Property PredictionProperty PredictionLearning Invariant Representations for Equivariant Neural Networks Using Orthogonal Moments
The convolutional layers of standard convolutional neural networks (CNNs) are equivariant to translation. However, the convolution and fully-connected layers are not equivariant or invariant to other affine geometric tra…
Rotated MNISTTranslation