Geometry-Aware Self-Training for Unsupervised Domain Adaptationon Object Point Clouds

TL;DR

This paper introduces GAST, a geometry-aware self-training approach for unsupervised domain adaptation of object point clouds, leveraging geometric tasks to improve cross-dataset classification accuracy.

Contribution

It proposes two novel self-supervised geometric tasks for learning domain-shared representations, addressing geometric variations in point cloud data.

Findings

GAST significantly outperforms state-of-the-art methods on PointDA-10.

The method effectively captures global topological configurations.

Normalization of point distribution improves domain adaptation.

Abstract

The point cloud representation of an object can have a large geometric variation in view of inconsistent data acquisition procedure, which thus leads to domain discrepancy due to diverse and uncontrollable shape representation cross datasets. To improve discrimination on unseen distribution of point-based geometries in a practical and feasible perspective, this paper proposes a new method of geometry-aware self-training (GAST) for unsupervised domain adaptation of object point cloud classification. Specifically, this paper aims to learn a domain-shared representation of semantic categories, via two novel self-supervised geometric learning tasks as feature regularization. On one hand, the representation learning is empowered by a linear mixup of point cloud samples with their self-generated rotation labels, to capture a global topological configuration of local geometries. On the other hand, a diverse point distribution across datasets can be normalized with a novel curvature-aware distortion localization. Experiments on the PointDA-10 dataset show that our GAST method can significantly outperform the state-of-the-art methods.