# PointSFDA: Source-free Domain Adaptation for Point Cloud Completion

arXiv: https://arxiv.org/abs/2503.15144 · HTML: https://mlsanity.com/paper/2503.15144/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Mar 2025.

## Built on by (most cited first, 1 of 1)

- [Source-Free Domain Adaptation for Geospatial Point Cloud Semantic Segmentation](https://mlsanity.com/paper/2601.08375/) (2026)

## Abstract (arXiv)

Conventional methods for point cloud completion, typically trained on synthetic datasets, face significant challenges when applied to out-of-distribution real-world scans. In this paper, we propose an effective yet simple source-free domain adaptation framework for point cloud completion, termed PointSFDA. Unlike unsupervised domain adaptation that reduces the domain gap by directly leveraging labeled source data, PointSFDA uses only a pretrained source model and unlabeled target data for adaptation, avoiding the need for inaccessible source data in practical scenarios. Being the first source-free domain adaptation architecture for point cloud completion, our method offers two core contributions. First, we introduce a coarse-to-fine distillation solution to explicitly transfer the global geometry knowledge learned from the source dataset. Second, as noise may be introduced due to domain gaps, we propose a self-supervised partial-mask consistency training strategy to learn local geometry information in the target domain. Extensive experiments have validated that our method significantly improves the performance of state-of-the-art networks in cross-domain shape completion. Our code is available at emphtextcolormagentahttps://github.com/Starak-x/PointSFDA.