The reformulate city-scale 3D grounding as structured constraint reasoning, where description semantics are organized into computable cross-modal constraints over open-vocabulary 3D entities, attributes, and spatial relations, and introduces CitySTAR-3D, an enhanced benchmark that improves semantic coverage, instance completeness, bounding-box fidelity, and spatial-relation complexity in city-scale 3D grounding.
Abstract
3D grounding aims to localize target entities in complex scenes from natural language and plays a fundamental role in embodied perception and spatial reasoning. However, existing approaches mostly rely on feature similarity or direct matching, making it difficult to connect natural-language intent with the implicit semantic and geometric structures hidden in billion-scale urban point clouds. We reformulate city-scale 3D grounding as structured constraint reasoning, where description semantics are organized into computable cross-modal constraints over open-vocabulary 3D entities, attributes, and spatial relations. We present CitySTAR, a training-free framework for reasoning-driven urban 3D grounding. CitySTAR lifts raw billion-scale urban point clouds into a query-ready scene graph of open-vocabulary 3D instances, with CodeLLM-driven tools supplying multimodal evidence for node attributes and 3D spatial relations. It then models target-context topology with paired hypergraphs and performs bidirectional topology verification for structural disambiguation. Finally, a Reflective Cross-modal Grounding module integrates topology consistency and candidate-centered 2D visual evidence to make decisions over a metric-aware 3D context graph. To further support this setting, we introduce CitySTAR-3D, an enhanced benchmark that improves semantic coverage, instance completeness, bounding-box fidelity, and spatial-relation complexity in city-scale 3D grounding. Extensive experiments show that CitySTAR consistently improves open-world urban 3D grounding while maintaining strong interpretability and generalization.
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-centr...
B. Nguyen, K. Nguyen, Clinton Fookes et al.· 0 citations
Semantic 3D city models offer rich urban information yet remain largely inaccessible to non-expert users due to their complex schemas, deep class hierarchies, and heterogeneous data structures. Large Language Models (LLMs) have demonstrated remarkable capabilities in natural language understanding and code generation....
K. Kanna, T. Kolbe· ISPRS Annals of the Photogra...· 0 citations
Urban spatial intelligence calls for 3D cities that agents can measure, interpret, and modify across successive tasks. Maintaining such cities as persistent, code-based states requires lightweight acquisition and support for consistent state updates without repeated dense reconstruction. We therefore formulate a new se...
Jia-Hui Sun, Ran Bi, Yi-Fei Liu et al.· 0 citations
A feed-forward, spatially constrained framework that grounds task-relevant objects and generates position-aware subtasks without an iterative task–scene refinement loop is proposed and confidence-guided verification improves the unweighted scene-average Relaxed Accuracy F1.
AnchorVLN is introduced, an open-vocabulary VLN system built on a simple rule: the VLM proposes semantics; geometry decides metrics, which enforces the semantic-geometry boundary without modifying the downstream autonomy stack.
Long Giang Vu, Cheng-Kai Yao, Yu-Xin Liu et al.· 0 citations
Large language models (LLMs) encode substantial latent geographic knowledge, yet they reason poorly over space and are unreliable when queried from coordinates alone. Useful behaviour emerges only when structured spatial context is supplied in the prompt. This raises a question geographic evaluation has left unexamined...
Joan Perez· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.