PC-Adapter: Topology-Aware Adapter for Efficient Domain Adaption on Point Clouds with Rectified Pseudo-label

TL;DR

This paper introduces PC-Adapter, a topology-aware adapter that efficiently adapts point cloud models across domains by preserving global geometry and learning local features, with a novel pseudo-labeling strategy to handle classifier bias.

Contribution

The paper presents a new adapter-based domain adaptation method for point clouds that leverages attention and graph convolution, along with a robust pseudo-labeling technique, addressing domain shift challenges.

Findings

Outperforms baseline methods on multiple benchmark datasets.

Effectively preserves global shape information during adaptation.

Demonstrates robustness to domain shifts in various scenarios.

Abstract

Understanding point clouds captured from the real-world is challenging due to shifts in data distribution caused by varying object scales, sensor angles, and self-occlusion. Prior works have addressed this issue by combining recent learning principles such as self-supervised learning, self-training, and adversarial training, which leads to significant computational overhead.Toward succinct yet powerful domain adaptation for point clouds, we revisit the unique challenges of point cloud data under domain shift scenarios and discover the importance of the global geometry of source data and trends of target pseudo-labels biased to the source label distribution. Motivated by our observations, we propose an adapter-guided domain adaptation method, PC-Adapter, that preserves the global shape information of the source domain using an attention-based adapter, while learning the local characteristics of the target domain via another adapter equipped with graph convolution. Additionally, we propose a novel pseudo-labeling strategy resilient to the classifier bias by adjusting confidence scores using their class-wise confidence distributions to consider relative confidences. Our method demonstrates superiority over baselines on various domain shift settings in benchmark datasets - PointDA, GraspNetPC, and PointSegDA.