arXiv：cs.LG· Jose Andres Millan-Romera, Samuel Cognolato, Holger Voos, Jose Luis Sanchez-Lopez, Luciano Serafini·· 4 小时前AI 评分33
通过自回归扩散模型在 3D 场景图中生成高层概念
Generation of High-Level Concepts in 3D Scene Graphs via Autoregressive Diffusion
AI 导读
研究者提出一种统一的自回归扩散图生成模型，联合学习图结构与空间节点特征，从观测到的垂直平面自底向上构建任意层级深度的完整 3D 场景图（3DSG）。该方法在合成场景、真实建筑平面图和机器人传感器数据等 3DSG 数据集上一致超越所有基于学习和随机的基线，并在最大层级和真实单层数据上超越可获取目标图规模信息的 one-shot 模型。
正文
Abstract:Indoor 3D Scene Graphs (3DSGs) represent environments as multi-layer hierarchies that connect observed geometric primitives (e.g., planes) to higher-level metric-semantic concepts (e.g., rooms, floors, buildings), enabling incremental spatial reasoning for robotic perception and SLAM. However, classical high-level concept generation approaches rely on hand-crafted rules for specific concept classes, while learning-based methods require separate models for graph structure and spatial node features (e.g., centroids), which limits scalability to novel classes and more complex hierarchies. We propose a unified autoregressive diffusion-based graph generative model that jointly learns structure and features, constructing complete 3DSGs bottom-up from observed vertical planes across arbitrary hierarchy depths. Our method consistently surpasses all learning-based and random baselines across 3DSG datasets spanning synthetic scenes, real architectural floor plans, and robotic sensor data, with varying layout complexity and hierarchy depth, and surpasses a one-shot model with oracle access to the target graph size on the largest hierarchy and on real single-floor data. Finally, we propose an adaptation of the Fused Gromov--Wasserstein distance for principled graph-level evaluation of generated 3DSGs against ground truth.
| Subjects: | Robotics (cs.RO); Machine Learning (cs.LG) |
| Cite as: | arXiv:2608.28733 [cs.RO] |
| (or arXiv:2608.28733v2 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2608.28733 arXiv-issued DOI via DataCite |
Submission history
From: José Andrés Millán Romera [view email]
[v1]
Fri, 28 Aug 2026 17:41:37 UTC (1,306 KB)
[v2]
Mon, 5 Oct 2026 18:54:53 UTC (7,410 KB)
来源：arXiv：cs.LG · arxiv.org