GSANet: Geometric Structure-Aware Siamese Network for 3D Change Detection.
Saved in:
| Title: | GSANet: Geometric Structure-Aware Siamese Network for 3D Change Detection. |
|---|---|
| Authors: | Chen, Jiakang1 (AUTHOR), Wang, Rongfang1,2 (AUTHOR) rfwang@xidian.edu.cn, Sun, Libin1,2 (AUTHOR), Jiao, Changzhe1,2 (AUTHOR) |
| Source: | Remote Sensing. Aug2026, Vol. 18 Issue 16, p2647. 21p. |
| Subjects: | Point cloud, Geometry, Artificial neural networks, Remote sensing |
| Abstract: | Highlights: What are the main findings? A Geometric Structure-Aware Siamese Network (GSANet) is proposed for 3D point cloud change detection, which explicitly integrates boundary and edge priors via Boundary-Aware Subsampling (BAS), Edge-Aware Adaptive Graph Convolution, and a Difference Enhancement Module (DEM). GSANet achieves state-of-the-art performance on the SLPCCD and HKCD datasets, with an mIoU of 82.87% and OA of 97.30%, outperforming existing methods including 3DCDNet, PGN3DCD, and Ms-DANet. What are the implications of the main findings? The proposed framework enhances feature representation robustness by preserving critical geometric cues, effectively suppressing false positives and improving boundary localization in complex urban environments affected by noise and viewpoint variations. It offers a practical and reliable solution for real-world remote sensing applications such as urban monitoring, disaster assessment, and environmental analysis, particularly in scenarios with sparse or occluded point cloud data. Three-dimensional (3D) point cloud change detection is essential for urban monitoring and environmental analysis, yet existing methods mainly rely on point-wise semantic differences and overlook change–unchange boundaries and object edges—key geometric cues for precise localization, especially for subtle or gradual changes. To address this, we propose GSANet, a Geometric Structure-Aware Siamese Network that explicitly integrates boundary and edge priors into sampling and feature learning. First, a Boundary-Aware Subsampling (BAS) strategy preserves key points near change boundaries while reducing redundancy, and a Boundary-Aware Binary Cross-Entropy (BA-BCE) loss assigns higher supervision weights to boundary points, enhancing learning in ambiguous regions. Second, an Edge-Aware Siamese Network captures robust local shapes by embedding edge priors into feature extraction, incorporating Edge-Aware Adaptive Graph Convolution, Edge-Aware Downsampling, and Cross-Attention Upsampling to maintain structural consistency across temporal branches. Additionally, a Difference Enhancement Module (DEM) amplifies feature discrepancies between bitemporal point clouds, improving sensitivity to subtle changes. Extensive experiments on a street-level dataset and urban dataset show our method outperforms state-of-the-art approaches. [ABSTRACT FROM AUTHOR] |
| Copyright of Remote Sensing is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Highlights: What are the main findings? A Geometric Structure-Aware Siamese Network (GSANet) is proposed for 3D point cloud change detection, which explicitly integrates boundary and edge priors via Boundary-Aware Subsampling (BAS), Edge-Aware Adaptive Graph Convolution, and a Difference Enhancement Module (DEM). GSANet achieves state-of-the-art performance on the SLPCCD and HKCD datasets, with an mIoU of 82.87% and OA of 97.30%, outperforming existing methods including 3DCDNet, PGN3DCD, and Ms-DANet. What are the implications of the main findings? The proposed framework enhances feature representation robustness by preserving critical geometric cues, effectively suppressing false positives and improving boundary localization in complex urban environments affected by noise and viewpoint variations. It offers a practical and reliable solution for real-world remote sensing applications such as urban monitoring, disaster assessment, and environmental analysis, particularly in scenarios with sparse or occluded point cloud data. Three-dimensional (3D) point cloud change detection is essential for urban monitoring and environmental analysis, yet existing methods mainly rely on point-wise semantic differences and overlook change–unchange boundaries and object edges—key geometric cues for precise localization, especially for subtle or gradual changes. To address this, we propose GSANet, a Geometric Structure-Aware Siamese Network that explicitly integrates boundary and edge priors into sampling and feature learning. First, a Boundary-Aware Subsampling (BAS) strategy preserves key points near change boundaries while reducing redundancy, and a Boundary-Aware Binary Cross-Entropy (BA-BCE) loss assigns higher supervision weights to boundary points, enhancing learning in ambiguous regions. Second, an Edge-Aware Siamese Network captures robust local shapes by embedding edge priors into feature extraction, incorporating Edge-Aware Adaptive Graph Convolution, Edge-Aware Downsampling, and Cross-Attention Upsampling to maintain structural consistency across temporal branches. Additionally, a Difference Enhancement Module (DEM) amplifies feature discrepancies between bitemporal point clouds, improving sensitivity to subtle changes. Extensive experiments on a street-level dataset and urban dataset show our method outperforms state-of-the-art approaches. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 20724292 |
| DOI: | 10.3390/rs18162647 |
