OccluGaussian: Occlusion-Aware Gaussian Splatting for Large Scene Reconstruction and Rendering

2025-03-20

AI Summary

In large-scale scene reconstruction, 3D Gaussian Splatting (3DGS) suffers from degraded camera correlation and reconstruction quality due to occlusion-agnostic scene partitioning. To address this, we propose an occlusion-aware region partitioning and rendering framework. Our method jointly models camera co-visibility and spatial geometry for hierarchical clustering—marking the first such integration—and introduces a visibility-driven Gaussian pruning mechanism that retains only those Gaussians jointly visible across multiple views within each region. Evaluated on multiple large real-world scenes, our approach achieves significant improvements in reconstruction accuracy while preserving rendering visual fidelity. Moreover, it accelerates rendering by over 40% compared to state-of-the-art methods, enabling scalable, high-fidelity, real-time reconstruction and rendering.

Technology Category

• Computer Vision: 3D Computer Vision
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Security and Privacy: Data transparency and provenance
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines

Abstract

In large-scale scene reconstruction using 3D Gaussian splatting, it is common to partition the scene into multiple smaller regions and reconstruct them individually. However, existing division methods are occlusion-agnostic, meaning that each region may contain areas with severe occlusions. As a result, the cameras within those regions are less correlated, leading to a low average contribution to the overall reconstruction. In this paper, we propose an occlusion-aware scene division strategy that clusters training cameras based on their positions and co-visibilities to acquire multiple regions. Cameras in such regions exhibit stronger correlations and a higher average contribution, facilitating high-quality scene reconstruction. We further propose a region-based rendering technique to accelerate large scene rendering, which culls Gaussians invisible to the region where the viewpoint is located. Such a technique significantly speeds up the rendering without compromising quality. Extensive experiments on multiple large scenes show that our method achieves superior reconstruction results with faster rendering speed compared to existing state-of-the-art approaches. Project page: https://occlugaussian.github.io.

Problem

• Occlusion-agnostic division reduces reconstruction quality in large scenes.
• Proposes occlusion-aware camera clustering for better scene reconstruction.
• Region-based rendering accelerates large scene rendering without quality loss.

Innovation

• Occlusion-aware scene division strategy
• Region-based rendering technique
• Culling invisible Gaussians for speed