geospatial data integration

Existing spatial pattern matching methods are largely confined to two-dimensional space and struggle to handle three-dimensional entity matching involving elevation or height information in real-world scenarios. This work extends spatial pattern matching to 3D environments for the first time, introducing a general problem formulation and proposing a subgraph-matching-based algorithm that explicitly models distance relationships in three-dimensional space. To support empirical evaluation, the authors construct the first 3D spatial pattern matching dataset, integrating both synthetic data and real-world building structures from the city of Hamburg. Experimental results on this benchmark demonstrate the effectiveness of the proposed approach, establishing a foundational algorithmic framework and experimental platform for future research in 3D spatial pattern analysis.

In heterogeneous data integration, schema matching and entity resolution are significantly affected by domain-specific characteristics, data scale, missingness rates, and attribute overlap—yet existing graph-based methods struggle to jointly leverage structural and semantic information. Method: This paper proposes a context-aware graph embedding framework that unifies tabular structure, column-level textual descriptions, and external knowledge, employing graph neural networks for joint encoding and embedding learning of multi-source heterogeneous data. Contribution/Results: A key innovation is the context-enhancement mechanism, which systematically uncovers how data characteristics influence matching performance and empirically demonstrates that contextual modeling substantially improves robustness and accuracy—especially under challenging conditions such as high missingness rates and prevalent numeric columns. Extensive experiments across multiple domain-specific benchmark datasets show that our method consistently outperforms state-of-the-art graph-based baselines.

Addressing the entity matching challenge in multi-source 3D geospatial data—particularly when spatial proximity, textual metadata, or external identifiers are missing, unreliable, or inconsistent—this paper proposes 3dSAGER, an end-to-end framework. Methodologically, it integrates deep geometric feature extraction, learning-based candidate generation, 3D shape encoding, and similarity matching, trained end-to-end on real-world urban datasets. Its key contributions are: (1) a novel coordinate-system-agnostic 3D geometric feature representation, derived solely from intrinsic object shape structure; and (2) BKAFI, a lightweight, interpretable blocking method that enhances cross-source matching robustness and efficiency. Experiments demonstrate that 3dSAGER significantly outperforms strong baselines in both accuracy and inference speed. Ablation studies confirm the effectiveness and design rationale of each component.

To address the misalignment of cross-platform 3D models caused by the absence of geospatial metadata in ground- and airborne imagery, this paper proposes a fully metadata-free georegistration method. Our approach leverages only publicly available satellite imagery and a digital surface model (DSM), integrating neural radiance fields (NeRF) modeling with differentiable rendering to achieve robust geolocalization of airborne imagery via multi-view geometric optimization, and subsequently enables precise alignment of ground-level imagery to the airborne 3D model. We introduce, for the first time, a novel “satellite + DSM + NeRF” collaborative registration paradigm, supporting unified georegistration of non-overlapping and disconnected 3D models. Evaluated across multiple real-world scenes, our method achieves an average geolocation error of under 5 meters—without requiring any raw sensor metadata such as GPS or IMU readings.

This paper identifies and systematically analyzes common challenges in spatial data science across mainstream programming languages—R, Python, and Julia—including inconsistent spherical geometry modeling, ambiguous spatial/temporal semantics, conflation of intensive and extensive attributes, poor interoperability between data cube and vector formats, complex cross-package dependencies, and a persistent divide between GIS and physical modeling communities. Through multi-language ecosystem surveys, cross-community comparative analysis, and software engineering abstraction, we propose, for the first time, a cross-language semantic framework for spatial operations. The framework formally defines support types (point vs. block), specifies attribute-type constraints on operation validity, and refactors spherical Simple Features logic. We distill five foundational insights that establish a methodological basis and practical guidance for tool interoperability, pedagogical alignment, and open-source governance in spatial computing.

This study addresses the challenges urban decision-making faces due to highly heterogeneous spatial data and the limited semantic reasoning capabilities of existing GIS tools, which often lead to error-prone manual processing. To overcome these limitations, this work proposes UrbanTrace, a novel visual analytics system that integrates semantic-aware large language model (LLM) agents into the spatial data integration pipeline for the first time. UrbanTrace employs an offline parser to extract semantic and geometric metadata, combines domain-customized LLM agents with a nodal workflow architecture, and enables goal-driven data discovery and legally compliant spatial aggregation. The system innovatively transforms spatial aggregation sensitivity into explorable visual assets and transparently reveals the integration process and outcomes through a tri-view interface. Evaluations across 28 urban scenarios and 112 datasets demonstrate 100% semantic validity in data discovery and 87% geometric validity in spatial mapping, with expert assessments confirming significant improvements in analytical reliability and exploratory efficiency.

This work addresses the limitations of existing Earth observation foundation models, which predominantly rely on raster data and overlook the structured geographic semantics embedded in open vector datasets such as OpenStreetMap, thereby hindering comprehensive understanding of human–environment systems. To overcome this, we propose the first unified spatial representation learning framework that deeply integrates remote sensing imagery and vector data within a shared embedding space, breaking away from conventional modality-isolated paradigms. By leveraging self-supervised learning and multimodal alignment—while explicitly modeling geometric, topological, and semantic relationships—our approach enables synergistic raster perception and vector-based reasoning. The method substantially enhances accuracy, semantic interpretability, and explainability on downstream tasks, laying a theoretical and methodological foundation for developing human-centered, semantically rich geospatial foundation models.

This work addresses the challenge of unified modeling for multi-class heterogeneous geographic entities in remote sensing vector mapping, where existing methods struggle to adequately represent topological relationships and instance boundaries. The authors propose reframing vector map construction as a structured text generation task by designing a GeoJSON-like hierarchical vector language that jointly encodes geometric, semantic, and topological information. A progressive vision-to-language mapping framework is introduced, optimized via reinforcement learning to ensure syntactic validity, content fidelity, and map executability of the generated output, thereby enabling cross-category unified modeling. Experiments on the newly curated VecMap-Bench dataset—comprising 54K images and 800K instances—demonstrate that the proposed approach significantly outperforms state-of-the-art methods in single- and multi-class mapping, cross-dataset transfer, and open-vocabulary generalization.

This study addresses the challenges in intelligent transportation research stemming from the absence of unified standards for multi-source heterogeneous data, inconsistent model interfaces, and non-standardized evaluation protocols, which collectively hinder reproducibility, cross-dataset benchmarking, and cross-regional transferability. To overcome these limitations, the authors propose a unified five-layer platform encompassing hardware, data, models, evaluation, and prototyping. The work introduces an object-oriented trajectory representation—incorporating oriented bounding boxes and kinematic variables—that enables, for the first time, a standardized format across diverse trajectory datasets. The platform integrates CARLA-based digital twin maps, calibrated traffic models, automated data conversion pipelines, and standardized evaluation protocols. Experimental results demonstrate an 85% reduction in experimental setup time, 91% efficiency in cross-city safety model transfer, and cross-dataset reproduction variance below 3%.

This work addresses the limitations of traditional database migration approaches, which are often constrained to specific source-target model pairs and struggle to support general-purpose migration in heterogeneous, multi-model environments. To overcome this, the authors propose a model-driven migration framework based on a unified data model called U-Schema. By mapping diverse data models into a common intermediate representation, the framework drastically reduces the number of required transformation pathways and enables cross-paradigm migrations. It employs traceable metadata to decouple schema transformation from data migration, thereby preserving semantic consistency while enhancing structural fidelity and query behavior equivalence. Empirical evaluation—conducted in a relational-to-document database migration scenario using both synthetic datasets and the Northwind benchmark—demonstrates the approach’s effectiveness and scalability across varying data sizes.