GS4Buildings: Prior-Guided Gaussian Splatting for 3D Building Reconstruction

2025-08-10

AI Summary

To address incomplete building surface reconstruction in complex urban scenes caused by occlusion in 2D Gaussian Splatting (2DGS), this paper proposes a semantic prior-guided Gaussian splatting method. We first integrate LoD2-level semantic 3D building models into the 2DGS framework to initialize geometrically consistent Gaussian primitives. We further fuse prior depth and normal maps derived from planar geometry and impose building-region masks to enhance optimization stability. Additionally, we design a building-focused optimization mode that significantly suppresses redundant primitives. Evaluated on urban datasets, our method improves reconstruction completeness by 20.5%, geometric accuracy by 32.8%, and reduces the number of Gaussian primitives by 71.8%. These gains collectively advance the completeness, fidelity, and compactness of large-scale urban building reconstruction.

Technology Category

• Computer Vision: 3D Computer Vision
• Search and Optimization: Sampling/Simulation-based Search
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Recent advances in Gaussian Splatting (GS) have demonstrated its effectiveness in photo-realistic rendering and 3D reconstruction. Among these, 2D Gaussian Splatting (2DGS) is particularly suitable for surface reconstruction due to its flattened Gaussian representation and integrated normal regularization. However, its performance often degrades in large-scale and complex urban scenes with frequent occlusions, leading to incomplete building reconstructions. We propose GS4Buildings, a novel prior-guided Gaussian Splatting method leveraging the ubiquity of semantic 3D building models for robust and scalable building surface reconstruction. Instead of relying on traditional Structure-from-Motion (SfM) pipelines, GS4Buildings initializes Gaussians directly from low-level Level of Detail (LoD)2 semantic 3D building models. Moreover, we generate prior depth and normal maps from the planar building geometry and incorporate them into the optimization process, providing strong geometric guidance for surface consistency and structural accuracy. We also introduce an optional building-focused mode that limits reconstruction to building regions, achieving a 71.8% reduction in Gaussian primitives and enabling a more efficient and compact representation. Experiments on urban datasets demonstrate that GS4Buildings improves reconstruction completeness by 20.5% and geometric accuracy by 32.8%. These results highlight the potential of semantic building model integration to advance GS-based reconstruction toward real-world urban applications such as smart cities and digital twins. Our project is available: https://github.com/zqlin0521/GS4Buildings.

Problem

• Improves 3D building reconstruction in complex urban scenes
• Uses semantic 3D models for robust surface reconstruction
• Reduces Gaussian primitives for efficient building representation

Innovation

• Prior-guided Gaussian Splatting for building reconstruction
• Initializes Gaussians from LoD2 semantic models
• Incorporates prior depth and normal maps