OpenSplatGraph: From Dense Semantic Maps to Structured Scene Graphs for Open-Vocabulary Robot Perception

Binh Long Nguyen; Kien Nguyen; Clinton Fookes; Peyman Moghadam

cs.ROOct 6, 2026

OpenSplatGraph: From Dense Semantic Maps to Structured Scene Graphs for Open-Vocabulary Robot Perception

Authors: Binh Long Nguyen, Kien Nguyen, Clinton Fookes, Peyman Moghadam

Organizations: School of Electrical Engineering and Robotics, Queensland University of Technology (QUT), Brisbane, QLD 4000, Australia · CSIRO Robotics, CSIRO, Brisbane, QLD 4069, Australia

Abstract

Dense 3D mapping with semantic understanding is essential for robotic perception in complex environments. Recent 3D Gaussian Splatting-based mapping approaches enable high-fidelity geometry and efficient open-vocabulary perception, but typically represent semantics as unstructured feature fields that limit object-centric reasoning. In contrast, 3D scene graphs explicitly model objects and their relationships for structured reasoning, but are commonly constructed from sparse geometric representations that do not fully exploit dense semantic maps. In this work, we present OpenSplatGraph, a unified framework that constructs persistent 3D scene graphs directly from an online Gaussian-based open-vocabulary semantic map. The proposed framework augments the dense semantic map with a reliability-aware semantic field that maintains lightweight observation statistics for confidence-aware, query-conditioned object extraction. Extracted object instances are associated with persistent graph nodes, allowing object attributes and relationships to be incrementally updated across observations and queries. By tightly coupling dense semantic mapping with persistent object-centric representations, our framework supports both language-guided object grounding and structured relational reasoning while preserving the geometric fidelity of Gaussian-based mapping. Comprehensive evaluations on standard 3D scene understanding benchmarks and real-world robotic experiments demonstrate that OpenSplatGraph achieves competitive performance for online open-vocabulary perception and downstream robotic tasks. Project page: https://csiro-robotics.github.io/OpenSplatGraph.

Figures & tables

Figure 1 : Comparison of complementary representations for open-vocabulary 3D scene understanding. Dense semantic maps provide high-fidelity geometry and language-aligned features but represent semantics as unstructured feature fields. Scene graphs explicitly model objects and their relationships for structured reasoning, but typically rely on sparse geometric representations. Our framework bridges these paradigms by constructing a persistent 3D scene graph directly from an online dense semantic map.

Figure 2 : Overview of OpenSplatGraph . Starting from a multi-view RGB-D sequence, the framework incrementally constructs an online open-vocabulary dense semantic map together with a reliability-aware semantic field for robust query-conditioned object extraction. Extracted object instances are then associated across queries and incrementally accumulated into a persistent 3D scene graph. By jointly maintaining the dense semantic field and the scene graph, OpenSplatGraph enables structured relational reasoning while preserving high-fidelity geometric and semantic information for downstream robotic tasks.

Mode	Method	ScanNet					Replica
Mode	Method	ScanNet20		NYUv2-40			ScanNet20		NYUv2-40
		mIoU $\uparrow$	mAcc $\uparrow$	mIoU $\uparrow$	mAcc $\uparrow$	FPS $\uparrow$	mIoU $\uparrow$	mAcc $\uparrow$	mIoU $\uparrow$	mAcc $\uparrow$	FPS $\uparrow$
Offline	LangSplat [ 23 ]	3.78	9.11	3.46	8.43	–	3.84	8.75	3.56	8.26	–
	OpenGaussian [ 33 ]	30.08	45.59	27.53	42.17	–	25.63	36.24	23.45	31.59	–
	InstanceGaussian [ 17 ]	40.32	55.78	36.88	52.91	–	32.24	45.99	29.88	41.01	–
Online	ConceptFusion [ 11 ]	12.80	22.62	11.72	20.93	0.52	13.05	29.89	11.51	27.66	0.49

Table 1 : Quantitative comparison of 3D semantic segmentation on ScanNet and Replica datasets. (Best and second-best results are highlighted in bold and underlined .)

Figure 3 : Qualitative results of 3D semantic segmentation on ScanNet and Replica datasets.

Query Type	CLIP Retrieval						LLM Retrieval
	ConceptGraphs			OpenSplatGraph			ConceptGraphs			OpenSplatGraph
	R@1	R@2	R@3	R@1	R@2	R@3	R@1	R@2	R@3	R@1	R@2	R@3
Descriptive	0.59	0.82	0.86	0.45	0.82	0.92	0.61	0.64	0.64	0.50	0.59	0.69
Affordance	0.43	0.57	0.63	0.54	0.61	0.79	0.57	0.63	0.66	0.61	0.71	0.75
Negation	0.26	0.60	0.71	0.51	0.64	0.82	0.80	0.89	0.97	0.80	0.89	0.94

Table 2 : Quantitative comparison of 3D object grounding on Replica across different query types and retrieval strategies. (ConceptGraphs results are reported from [ 7 ] .)

Figure 4 : Qualitative visualization of open-vocabulary 3D scene graph construction on the Replica dataset. For visualization, asymmetric relations are displayed from nodes with lower instance IDs to those with higher IDs.

Figure 5 : Real-world robotic navigation experiments with OpenSplatGraph . The framework constructs an open-vocabulary semantic map online and supports language-guided navigation in indoor environments. For each query, we visualize the rendered target object from the reference view and from an additional arbitrary viewpoint.

Configuration	mIoU $\uparrow$	mAcc $\uparrow$	R@1 $\uparrow$
Baseline	30.97	42.39	–
+ Reliability-aware semantic modeling	36.06	50.74	–
+ Persistent object association	36.06	50.74	0.36
+ Scene graph reasoning (Full)	36.06	50.74	0.50

Table 3 : Progressive ablation of OpenSplatGraph components.

Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

Method	ScanNet		Replica
Method	mIoU $\uparrow$	mAcc $\uparrow$	mIoU $\uparrow$	mAcc $\uparrow$
GaussianGraph [ 30 ]	31.09	48.91	31.18	49.14
DynamicGSG [ 6 ]	–	–	31.06 †	54.04 †
OpenSplatGraph (Ours)	36.75	53.82	35.37	47.67

Appendix

Table 4 : Additional open-vocabulary 3D semantic segmentation results on ScanNet and Replica. † DynamicGSG uses a different evaluation mapping.

Method	Node Prec. $\uparrow$	Edge Prec. $\uparrow$	Dup. Rate (%) $\downarrow$
ConceptGraphs [ 7 ]	0.71	0.88	6.77
OpenSplatGraph (Ours)	0.73	0.91	5.94

Appendix

Table 5 : Direct scene-graph evaluation against ConceptGraphs.

Explore similar work

LEXI-SG: Monocular 3D Scene Graph Mapping with Room-Guided Feed-Forward Reconstruction
May 13, 2026Christina Kassab, Hyeonjae Gil, Matías Mattamala +23D Scene Graphs Simultaneous Localization And Mapping
Department of Engineering Science, University of Oxford, UK · Department of Mechanical Engineering, Seoul National University, South Korea · School of Informatics, University of Edinburgh, UK
Think While You Map: Asynchronous Vision-Language Agents for Incremental 3D Scene Graphs
Jun 30, 2026Deniz Bickici, Michael Pabst, Shohei Mori +13D Scene Graphs Voxel
University of Stuttgart, Germany · IMPRS-IS, Germany · Graz University of Technology, Austria
DGSG-Mind: Dynamic 3D Gaussian Scene Graphs for Long-Term Scene Understanding and Grounding
May 28, 2026Luzhou Ge, Xiangyu Zhu, Jinyan Liu +13D Scene Graphs 3D Gaussian
School of Computer Science, Beijing Institute of Technology, Beijing 100081, China.

May 13, 2026, cs.ROJ/K move · Enter open · S save

LEXI-SG: Monocular 3D Scene Graph Mapping with Room-Guided Feed-Forward Reconstruction

Christina Kassab, Hyeonjae Gil, Matías Mattamala, Ayoung Kim+1

Department of Engineering Science, University of Oxford, UK · Department of Mechanical Engineering, Seoul National University, South Korea · School of Informatics, University of Edinburgh, UK

3D Scene Graphs Simultaneous Localization And Mapping Scene Understanding

Open paper AskPDF arXiv

Scene graphs are becoming a standard representation for robot navigation, providing hierarchical geometric and semantic scene understanding. However, most scene graph mapping methods rely on depth cameras or LiDAR sensors. In this work, we present LEXI-SG, the first dense monocular visual mapping system for open-vocabulary 3D scene graphs using only RGB camera input. Our approach exploits the semantic priors of open-vocabulary foundation models to partition the scene into rooms, deferring feed-forward reconstruction to when each room is fully observed -- enabling scalable dense mapping without sliding-window scale inconsistencies. We propose a room-based factor graph formulation to globally align room reconstructions while preserving local map consistency and naturally imposing the semantic scene graph hierarchy. Within each room, we further support open-vocabulary object segmentation and tracking. We validate LEXI-SG on indoor scenes from the Habitat-Matterport 3D and self-collected egocentric office sequences. We evaluate its performance against existing feed-forward SLAM methods, as well as established scene graphs baselines. We demonstrate improved trajectory estimation and dense reconstruction, as well as, competitive performance in open-vocabulary segmentation. LEXI-SG shows that accurate, scalable, open-vocabulary 3D scene graphs can be achieved from monocular RGB alone. Our project page and office sequences are available here: https://ori-drs.github.io/lexisg-web/.

Abstract

Explore similar work

LEXI-SG: Monocular 3D Scene Graph Mapping with Room-Guided Feed-Forward Reconstruction

Think While You Map: Asynchronous Vision-Language Agents for Incremental 3D Scene Graphs

DGSG-Mind: Dynamic 3D Gaussian Scene Graphs for Long-Term Scene Understanding and Grounding