Graph-GSReg: Leveraging 3D Scene Graphs for Gaussian Splatting Registration
2026-06-29 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors address the challenge of combining multiple 3D Gaussian Splatting scenes into one seamless 3D map, which is important for large-scale mapping. Instead of matching individual points, they represent each scene as a 3D scene graph, turning the problem into graph registration for better accuracy. They also introduce a way to fine-tune the merged scene without extra data to fix common issues like visual glitches. Their tests show that this method produces well-aligned and visually consistent merged scenes on different datasets.
3D Gaussian Splatting3D scene graphgraph registrationself-supervised optimizationscene mergingvisual consistencyregistration accuracy3D mappingocclusion artifacts
Authors
Jaewon Lee, Mangyu Kong, Euntai Kim
Abstract
Merging multiple 3D Gaussian Splatting (3DGS) scenes into a single unified Gaussian representation is essential for large-scale 3D mapping and long-term map management. Despite its importance, this area remains underexplored, and existing solutions exhibit several limitations. Learning-based methods attempt direct correspondence between Gaussian primitives and require training on large 3DGS datasets. Image-based optimization methods depend heavily on coarse initialization from generic foundation models and often incur expensive refinement. We present \ourmodel. Our method constructs a 3D scene graph from a 3DGS and its rendered images, \textit{reformulating 3DGS registration as a graph registration problem}. The proposed 3D scene graph represents each 3DGS at a higher-level representation, enabling a globally consistent understanding of semantic information and structural context for accurate registration. To further construct a seamless unified scene, we introduce a Self-Supervised Test-Time Optimization. Naively merging two 3D Gaussian scenes often suffers from occlusion artifacts such as hollows and floaters. To alleviate this issue, we refine the merged Gaussians to preserve visual consistency between the original scenes and the merged scene. We evaluate our method on real and synthetic benchmarks, demonstrating competitive registration accuracy and merged scene rendering quality.