Pointwise Rotation-Invariant Network with Adaptive Sampling and 3D Spherical Voxel Convolution

By Yang You, Yujing Lou, Qi Liu, Yu-Wing Tai, Lizhuang Ma, Cewu Lu, Weiming Wang. 2018

75 citations recorded by the source corpus.

Point cloud analysis without pose priors is very challenging in real applications, as the orientations of point clouds are often unknown. In this paper, we propose a brand new point-set learning framework PRIN, namely, Pointwise Rotation-Invariant Network, focusing on rotation-invariant feature extraction in point clouds analysis. We construct spherical signals by Density Aware Adaptive Sampling to deal with distorted point distributions in spherical space. In addition, we propose Spherical Voxel Convolution and Point Re-sampling to extract rotation-invariant features for each point. Our network can be applied to tasks ranging from object classification, part segmentation, to 3D feature matching and label alignment. We show that, on the dataset with randomly rotated point clouds, PRIN demonstrates better performance than state-of-the-art methods without any data augmentation. We also provide theoretical analysis for the rotation-invariance achieved by our methods.

Subfields: Computational Mechanics, Computer Vision and Pattern Recognition, Geology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Mechanics

1. Max Welling
2. Fabian B. Fuchs
3. Volker Fischer
4. Daniel E. Worrall
5. Bo Yang
6. Yulan Guo
7. Andrew Markham
8. Sheng Ao

In Computer Vision and Pattern Recognition

1. Daniel E. Worrall
2. Max Welling
3. Fabian B. Fuchs
4. Andrew Markham
5. Volker Fischer
6. Yulan Guo
7. Qingyong Hu
8. Bo Yang

In Geology

1. Max Welling
2. Bo Yang
3. Volker Fischer
4. Yulan Guo
5. Andrew Markham
6. Daniel E. Worrall
7. Fabian B. Fuchs
8. Sheng Ao

Citations in the corpus

• SE(3)-Transformers: 3D Roto-Translation Equivariant Attention Networks (2020, methodology): “Our method sets itself apart by using roto-translation equivariant layers acting directly on the point cloud without prior projection onto a sphere [22, 45, 7].”
• SpinNet: Learning a General Surface Descriptor for 3D Point Cloud Registration (2020, methodology): “Note that, the existing methods [50, 62] usually use handcrafted features to achieve rotation invariance, resulting in the loss of the rich local patterns.”