SAM-Based Building Change Detection with Distribution-Aware Fourier Adaptation and Edge-Constrained Warping

2025-04-17

AI Summary

To address domain shift, long-tailed building distribution, and bi-temporal misalignment in remote sensing building change detection using the Segment Anything Model (SAM), this paper proposes a Distribution-Aware Fourier Aggregation Adapter and an Edge-Constrained Deformation Module. The former mitigates domain discrepancy and class imbalance via frequency-domain adaptation and distribution-weighted feature aggregation; the latter enhances edge localization accuracy and suppresses background noise by integrating edge-guided optical flow optimization with deformation alignment. Built upon SAM’s encoder, our end-to-end framework achieves state-of-the-art performance on LEVIR-CD, S2Looking, and WHU-CD benchmarks. It significantly improves detection of small-scale changes (+4.2% ΔF1) and boundary localization accuracy (+5.8% mIoU), demonstrating robustness to domain shifts and class imbalance while effectively handling temporal misalignment.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Transfer, Domain Adaptation, Multi-Task Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Building change detection remains challenging for urban development, disaster assessment, and military reconnaissance. While foundation models like Segment Anything Model (SAM) show strong segmentation capabilities, SAM is limited in the task of building change detection due to domain gap issues. Existing adapter-based fine-tuning approaches face challenges with imbalanced building distribution, resulting in poor detection of subtle changes and inaccurate edge extraction. Additionally, bi-temporal misalignment in change detection, typically addressed by optical flow, remains vulnerable to background noises. This affects the detection of building changes and compromises both detection accuracy and edge recognition. To tackle these challenges, we propose a new SAM-Based Network with Distribution-Aware Fourier Adaptation and Edge-Constrained Warping (FAEWNet) for building change detection. FAEWNet utilizes the SAM encoder to extract rich visual features from remote sensing images. To guide SAM in focusing on specific ground objects in remote sensing scenes, we propose a Distribution-Aware Fourier Aggregated Adapter to aggregate task-oriented changed information. This adapter not only effectively addresses the domain gap issue, but also pays attention to the distribution of changed buildings. Furthermore, to mitigate noise interference and misalignment in height offset estimation, we design a novel flow module that refines building edge extraction and enhances the perception of changed buildings. Our state-of-the-art results on the LEVIR-CD, S2Looking and WHU-CD datasets highlight the effectiveness of FAEWNet. The code is available at https://github.com/SUPERMAN123000/FAEWNet.

Problem

• Address domain gap in SAM for building change detection
• Improve detection of subtle changes and edge extraction
• Reduce noise interference and misalignment in bi-temporal data

Innovation

• SAM encoder extracts remote sensing features
• Fourier adapter aggregates task-oriented information
• Edge-constrained warping refines building extraction