Search papers, labs, and topics across Lattice.
This paper introduces STARS-GS, a structure-aware 3D Gaussian Splatting framework designed to enhance large-scale aerial surface reconstruction by addressing key challenges in scene partitioning, geometric constraints, and regularization. By implementing a novel scene partitioning strategy that preserves continuous structures, a neighborhood-aware Gaussian organization that enhances local geometric conformity, and an adaptive regularization approach tailored to local characteristics, STARS-GS significantly improves reconstruction quality. Experimental results show a notable increase in the average F1-score from 0.640 to 0.698, highlighting its effectiveness in achieving greater geometric accuracy and surface completeness in complex scenes.
STARS-GS boosts aerial surface reconstruction accuracy by over 9% through innovative structure-aware techniques that tackle common pitfalls in existing methods.
Large-scale 3D surface reconstruction from aerial imagery is fundamental to geospatial mapping and urban modeling. Recent advances in 3D Gaussian Splatting (3DGS) have demonstrated considerable potential for this task. However, existing methods still face three major challenges in large and complex scenes: scene partitioning may split continuous scene elements across independently optimized sub-regions; geometric constraints mainly focus on the attributes of individual Gaussians while overlooking their local organization; and uniform regularization struggles to accommodate heterogeneous geometric structures. To address these issues, we propose STARS-GS, a structure-aware 3DGS framework for large-scale surface reconstruction. First, we introduce a structure-aware scene partitioning strategy that better preserves continuous scene structures during partitioning and reduces cross-region geometric inconsistencies and stitching artifacts through boundary refinement. Second, we develop neighborhood-aware Gaussian organization that extends geometric constraints from individual primitives to their neighborhood organization, encouraging Gaussians to better conform to local surface geometry. Third, we introduce adaptive surface regularization that adjusts the regularization strength according to local geometric characteristics, promoting geometric consistency in structured regions while preserving plausible variations in unstructured regions. Extensive experiments on large-scale aerial photogrammetry benchmarks demonstrate that STARS-GS consistently outperforms the evaluated Gaussian-based methods in surface reconstruction. It increases the average F1-score from 0.640 for the second-best method to 0.698, corresponding to a relative improvement of approximately 9.1\%, demonstrating effective improvements in geometric accuracy and surface completeness.
