English

Canonical Capsules: Self-Supervised Capsules in Canonical Pose

Computer Vision and Pattern Recognition 2021-11-29 v2 Machine Learning

Abstract

We propose a self-supervised capsule architecture for 3D point clouds. We compute capsule decompositions of objects through permutation-equivariant attention, and self-supervise the process by training with pairs of randomly rotated objects. Our key idea is to aggregate the attention masks into semantic keypoints, and use these to supervise a decomposition that satisfies the capsule invariance/equivariance properties. This not only enables the training of a semantically consistent decomposition, but also allows us to learn a canonicalization operation that enables object-centric reasoning. To train our neural network we require neither classification labels nor manually-aligned training datasets. Yet, by learning an object-centric representation in a self-supervised manner, our method outperforms the state-of-the-art on 3D point cloud reconstruction, canonicalization, and unsupervised classification.

Keywords

Cite

@article{arxiv.2012.04718,
  title  = {Canonical Capsules: Self-Supervised Capsules in Canonical Pose},
  author = {Weiwei Sun and Andrea Tagliasacchi and Boyang Deng and Sara Sabour and Soroosh Yazdani and Geoffrey Hinton and Kwang Moo Yi},
  journal= {arXiv preprint arXiv:2012.04718},
  year   = {2021}
}

Comments

NeurIPS 2021; The first two authors contributed equally; Project website: https://canonical-capsules.github.io

R2 v1 2026-06-23T20:49:43.968Z