Robust and Efficient 3D Gaussian Splatting for Urban Scene Reconstruction

2025-07-30

AI Summary

Addressing the challenge of simultaneously achieving real-time performance, high reconstruction quality, and multi-view appearance robustness in city-scale 3D Gaussian splatting reconstruction, this paper proposes an efficient end-to-end framework. Methodologically: (1) we introduce a visibility-guided image selection strategy coupled with adaptive density control to explicitly optimize Gaussian sparsity and hierarchical detail representation; (2) an appearance transformation module is incorporated to mitigate inter-view illumination and exposure inconsistencies; (3) we jointly enforce depth- and scale-aware regularization, anti-aliasing enhancement, and scene segmentation priors. Experiments demonstrate that our method achieves superior reconstruction accuracy, geometric fidelity, and appearance consistency compared to existing state-of-the-art approaches—while maintaining real-time rendering frame rates. To the best of our knowledge, this is the first work enabling high-fidelity, robust, end-to-end real-time reconstruction and rendering for large-scale urban scenes.

Technology Category

• Computer Vision: 3D Computer Vision
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

We present a framework that enables fast reconstruction and real-time rendering of urban-scale scenes while maintaining robustness against appearance variations across multi-view captures. Our approach begins with scene partitioning for parallel training, employing a visibility-based image selection strategy to optimize training efficiency. A controllable level-of-detail (LOD) strategy explicitly regulates Gaussian density under a user-defined budget, enabling efficient training and rendering while maintaining high visual fidelity. The appearance transformation module mitigates the negative effects of appearance inconsistencies across images while enabling flexible adjustments. Additionally, we utilize enhancement modules, such as depth regularization, scale regularization, and antialiasing, to improve reconstruction fidelity. Experimental results demonstrate that our method effectively reconstructs urban-scale scenes and outperforms previous approaches in both efficiency and quality. The source code is available at: https://yzslab.github.io/REUrbanGS.

Problem

• Fast reconstruction and real-time rendering of urban scenes
• Robustness against appearance variations in multi-view captures
• Controllable level-of-detail for efficient training and rendering

Innovation

• Scene partitioning for parallel training
• Controllable LOD strategy regulates Gaussian density
• Appearance transformation mitigates inconsistencies