Timezone: »

 
SeLCA: Self-Supervised Learning of Canonical Axis
Seungwook Kim · Yoonwoo Jeong · Chunghyun Park · Jaesik Park · Minsu Cho

@
in Symmetry and Geometry in Neural Representations (NeurReps) »
Event URL: https://openreview.net/forum?id=Lwp_CgpFuA0 »

Robustness to rotation is critical for point cloud understanding tasks, as point cloud features can be affected dramatically with respect to prevalent rotation changes. In this work, we propose SeLCA, a novel self-supervised learning framework to learn to the canonical axis of point clouds in a probabilistic manner. In essence, we propose to \textit{learn} rotational-equivariance by predicting the canonical axis of point clouds, and achieve rotational-invariance by aligning the point clouds using their predicted canonical axis. When integrated into a rotation-sensitive pipeline, SeLCA achieves competitive performances on the ModelNet40 classification task under unseen rotations. Most interestingly, our proposed method also shows high robustness to various real-world point cloud corruptions presented by the ModelNet40-C dataset, compared to the state-of-the-art rotation-invariant method.

Keywords: Rotation Invariance · Point cloud canonical axis · Point cloud understanding · Rotation Equivariance · SO(3)-equivariance

Author Information

Seungwook Kim (POSTECH)
Yoonwoo Jeong (POSTECH)
Chunghyun Park (POSTECH)
Jaesik Park (POSTECH)
Minsu Cho (POSTECH)

More from the Same Authors