Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point clouds

Nathaniel Thomas, Tess Smidt, Steven Kearnes, Lusann Yang, Li Li +2 more
2/22/2018
cs.LGcs.AIcs.CV

Official code: tensorfieldnetworks/tensorfieldnetworks (163 stars, MIT, last commit 2020-01-07).

Abstract

We introduce tensor field neural networks, which are locally equivariant to 3D rotations, translations, and permutations of points at every layer. 3D rotation equivariance removes the need for data augmentation to identify features in arbitrary orientations. Our network uses filters built from spherical harmonics; due to the mathematical consequences of this filter choice, each layer accepts as input (and guarantees as output) scalars, vectors, and higher-order tensors, in the geometric sense of these terms. We demonstrate the capabilities of tensor field networks with tasks in geometry, physics, and chemistry.

View on arXivView PDF

Code Implementations(1)

tensorfieldnetworks/tensorfieldnetworksOfficial100%
MIT
16333Python, Jupyter NotebookFeb 9, 20186 years agoMIT

Similar Papers

A Complete Guide to Spherical Equivariant Graph Transformers

85% match2025

SO(3)-Equivariant Neural Networks for Learning Vector Fields on Spheres

85% match2025

Vector Neurons: A General Framework for SO(3)-Equivariant Networks

85% match1 repo5292021

Reducing SO(3) Convolutions to SO(2) for Efficient Equivariant GNNs

85% match2023

A quatum inspired neural network for geometric modeling

83% match2024

TensorNet: Cartesian Tensor Representations for Efficient Learning of Molecular Potentials

82% match1 repo4822023

Cite this paper

@article{thomas2018tensor,
  title  = {Tensor field networks: Rotation- and translation-equivariant neural networks for 3D point clouds},
  author = {Nathaniel Thomas and Tess Smidt and Steven Kearnes and Lusann Yang and Li Li and Kai Kohlhoff and Patrick Riley},
  year   = {2018},
  eprint = {1802.08219},
  archivePrefix = {arXiv},
  url    = {https://arxiv.org/abs/1802.08219}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/1802.08219.svg)](https://codeofpaper.com/paper/1802.08219)

Discussion