Source-Free Domain Adaptation for Geospatial Point Cloud Semantic Segmentation

2026-01-13

AI Summary

This work addresses the domain shift in geospatial point cloud semantic segmentation caused by regional variations and diverse acquisition strategies. Under the challenging source-free setting—where access to source domain data is unavailable—the proposed LoGo framework achieves effective adaptation using only a pre-trained model and unlabeled target domain data. LoGo innovatively integrates class-balanced prototype estimation with optimal transport-based global distribution alignment. It further introduces a local-global dual-consistency pseudo-label filtering mechanism, combining locally independent anchor mining and multi-augmentation ensemble prediction to substantially mitigate feature collapse and category bias under long-tailed distributions. Experiments demonstrate that LoGo significantly outperforms existing source-free domain adaptation methods across multiple geospatial point cloud datasets, notably improving segmentation performance on tail classes and enhancing overall robustness.

Technology Category

• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Computer Vision: Remote Sensing / Geospatial AI
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Security and Privacy: Data transparency and provenance

Abstract

Semantic segmentation of 3D geospatial point clouds is pivotal for remote sensing applications. However, variations in geographic patterns across regions and data acquisition strategies induce significant domain shifts, severely degrading the performance of deployed models. Existing domain adaptation methods typically rely on access to source-domain data. However, this requirement is rarely met due to data privacy concerns, regulatory policies, and data transmission limitations. This motivates the largely underexplored setting of source-free unsupervised domain adaptation (SFUDA), where only a pretrained model and unlabeled target-domain data are available. In this paper, we propose LoGo (Local-Global Dual-Consensus), a novel SFUDA framework specifically designed for geospatial point clouds. At the local level, we introduce a class-balanced prototype estimation module that abandons conventional global threshold filtering in favor of an intra-class independent anchor mining strategy. This ensures that robust feature prototypes can be generated even for sample-scarce tail classes, effectively mitigating the feature collapse caused by long-tailed distributions. At the global level, we introduce an optimal transport-based global distribution alignment module that formulates pseudo-label assignment as a global optimization problem. By enforcing global distribution constraints, this module effectively corrects the over-dominance of head classes inherent in local greedy assignments, preventing model predictions from being severely biased towards majority classes. Finally, we propose a dual-consistency pseudo-label filtering mechanism. This strategy retains only high-confidence pseudo-labels where local multi-augmented ensemble predictions align with global optimal transport assignments for self-training.

Problem

• Source-Free Domain Adaptation
• Geospatial Point Cloud
• Semantic Segmentation
• Domain Shift
• Unsupervised Domain Adaptation

Innovation

• Source-Free Domain Adaptation
• Geospatial Point Cloud
• Local-Global Consistency
• Optimal Transport
• Long-Tailed Distribution