Self-Supervised Learning for Domain Adaptation on Point-Clouds

TL;DR

This paper explores the use of self-supervised learning for domain adaptation in 3D point cloud perception, introducing new pretext tasks and training methods that significantly improve performance.

Contribution

It is the first to apply SSL to domain adaptation in 3D point clouds, proposing Deformation Reconstruction and Point Cloud Mixup for better transfer learning.

Findings

Large improvement over baseline methods in classification tasks.

Effective domain adaptation for segmentation tasks.

Demonstrates the potential of SSL in 3D perception domain.

Abstract

Self-supervised learning (SSL) is a technique for learning useful representations from unlabeled data. It has been applied effectively to domain adaptation (DA) on images and videos. It is still unknown if and how it can be leveraged for domain adaptation in 3D perception problems. Here we describe the first study of SSL for DA on point clouds. We introduce a new family of pretext tasks, Deformation Reconstruction, inspired by the deformations encountered in sim-to-real transformations. In addition, we propose a novel training procedure for labeled point cloud data motivated by the MixUp method called Point cloud Mixup (PCM). Evaluations on domain adaptations datasets for classification and segmentation, demonstrate a large improvement over existing and baseline methods.