CLIPoint3D: Language-Grounded Few-Shot Unsupervised 3D Point Cloud Domain Adaptation

TL;DR

CLIPoint3D introduces a novel few-shot unsupervised domain adaptation framework for 3D point clouds, leveraging CLIP's language-vision capabilities with efficient fine-tuning and geometric cues.

Contribution

It is the first to adapt CLIP for 3D point cloud domain adaptation using knowledge-driven prompt tuning and lightweight fine-tuning strategies.

Findings

Achieves 3-16% accuracy improvements on PointDA-10 and GraspNetPC-10 benchmarks.

Utilizes entropy-guided view sampling for confident projection selection.

Employs optimal transport and prototype alignment losses for effective domain bridging.

Abstract

Recent vision-language models (VLMs) such as CLIP demonstrate impressive cross-modal reasoning, extending beyond images to 3D perception. Yet, these models remain fragile under domain shifts, especially when adapting from synthetic to real-world point clouds. Conventional 3D domain adaptation approaches rely on heavy trainable encoders, yielding strong accuracy but at the cost of efficiency. We introduce CLIPoint3D, the first framework for few-shot unsupervised 3D point cloud domain adaptation built upon CLIP. Our approach projects 3D samples into multiple depth maps and exploits the frozen CLIP backbone, refined through a knowledge-driven prompt tuning scheme that integrates high-level language priors with geometric cues from a lightweight 3D encoder. To adapt task-specific features effectively, we apply parameter-efficient fine-tuning to CLIP's encoders and design an entropy-guided view sampling strategy for selecting confident projections. Furthermore, an optimal transport-based alignment loss and an uncertainty-aware prototype alignment loss collaboratively bridge source-target distribution gaps while maintaining class separability. Extensive experiments on PointDA-10 and GraspNetPC-10 benchmarks show that CLIPoint3D achieves consistent 3-16% accuracy gains over both CLIP-based and conventional encoder-based baselines. Project page: https://sarthakm320.github.io/CLIPoint3D.