Spatial Representation Learning Beyond Pixels: Unifying Raster Data and Vector Semantics for Human-Centric Geospatial Foundation Models

• URL: http://arxiv.org/abs/2606.02374v1
• Date: Mon, 01 Jun 2026 15:21:58 GMT
• Title: Spatial Representation Learning Beyond Pixels: Unifying Raster Data and Vector Semantics for Human-Centric Geospatial Foundation Models
• Abstract summary: Recent advances in self-supervised learning have given rise to Earth Observation Foundation Models (EOFMs)<n>This paper calls for a paradigm shift toward joint Spatial Representation Learning (SRL) in an unified embedding space that integrate perception with vector-based reasoning.<n>We contend that such integration is essential for developing next-generation geospatial AI systems capable of more accurate, interpretable, and grounded understanding of the Earth.
• Score: 12.200359585933766
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Earth Observation (EO) has fundamentally transformed the monitoring of environmental processes and human activities up to planetary scale. Recent advances in self-supervised learning have given rise to Earth Observation Foundation Models (EOFMs), which leverage petabyte-scale unlabeled EO data to learn transferable representations across a wide range of downstream geospatial tasks. Despite these advances, current EOFMs remain largely confined to raster modalities, overlooking the rich, structured information encoded in openly-accessible vector data sources such as OpenStreetMap and Overture. Vector data provides explicit and compact representations of geographic entities, including geometry, topology, and semantic relationships, offering critical contextual signals that are often ambiguous or inaccessible in imagery alone. Raster and vector data thus represent complementary views of geographic space: raster data captures continuous physical and spectral patterns, while vector data encodes discrete objects and their relational structure and often represents more of the human rather than the physical systems (e.g. social or demographic data). However, existing geospatial representation learning paradigms treat these modalities in isolation, relying on imperfect and often lossy transformations to bridge them. This perspective paper calls for a paradigm shift toward joint Spatial Representation Learning (SRL) in an unified embedding space that integrate raster perception with vector-based reasoning. Building on emerging efforts in multimodal geospatial learning, we highlight conceptual foundations, technical challenges, and promising directions for aligning heterogeneous spatial data sources. We contend that such integration is essential for developing next-generation geospatial AI systems capable of more accurate, interpretable, and semantically grounded understanding of the Earth.

Related papers

• Textual Supervision Enhances Geospatial Representations in Vision-Language Models [47.5986642837928]
  Geospatial understanding is a critical yet underexplored dimension in the development of machine learning systems.<n>We analyze the geospatial representations acquired by three model families: vision-only architectures (e.g., ViT), vision-language models (e.g., CLIP), and large-scale multimodal foundation models (e.g., LLaVA, Qwen, and Gemma)
  arXiv  Detail & Related papers  (2026-06-05T11:40:13Z)
• GeoMamba: A Geometry-driven MambaVision Framework and Dataset for Fine-grained Optical-SAR Object Retrieval [54.741349848771144]
  GeoMamba is a geometry-driven framework for optical-SAR fine-grained retrieval.<n>GFI module enhances cross-modal feature interaction and incorporates structural priors.<n>GeoMamba outperforms existing methods, achieving 63.3% mAP and 77.0% Rank-1 accuracy in all-to-all retrieval setting.
  arXiv  Detail & Related papers  (2026-05-19T12:08:09Z)
• InfoGeo: Information-Theoretic Object-Centric Learning for Cross-View Generalizable UAV Geo-Localization [12.259964756337892]
  Cross-view geo-localization is fundamental for precise localization and navigation in GPS-denied environments.<n>We propose InfoGeo, an information-theoretic framework designed to enhance robustness and generalization.<n> InfoGeo reformulates the optimization as an information bottleneck process with two core objectives: (i) maximizing view-invariant information by aligning the object-centric structural relations across views, and (ii) minimizing view-specific noisy signals through cross-view knowledge constraints.
  arXiv  Detail & Related papers  (2026-05-08T01:28:49Z)
• GeoMeld: Toward Semantically Grounded Foundation Models for Remote Sensing [33.72621953693859]
  GeoMeld is a large-scale multimodal dataset with approximately 2.5 million spatially aligned samples.<n>GeoMeld provides semantically grounded language supervision through an agentic captioning framework.<n>GeoMeld-FM is a pretraining framework that combines multi-pretext masked autoencoding over aligned modalities, JEPA representation learning, and caption-vision contrastive alignment.
  arXiv  Detail & Related papers  (2026-04-12T11:47:11Z)
• GeoAgent: Learning to Geolocate Everywhere with Reinforced Geographic Characteristics [91.17301794848025]
  This paper presents GeoAgent, a model capable of reasoning closely with humans and deriving fine-grained address conclusions.<n>Previous RL-based methods have achieved breakthroughs in performance and interpretability but still remain concerns because of their reliance on AI-generated chain-of-thought (CoT) data and training strategies.
  arXiv  Detail & Related papers  (2026-02-13T04:48:05Z)
• GeoGNN: Quantifying and Mitigating Semantic Drift in Text-Attributed Graphs [59.61242815508687]
  Graph neural networks (GNNs) on text--attributed graphs (TAGs) encode node texts using pretrained language models (PLMs) and propagate these embeddings through linear neighborhood aggregation.<n>This work introduces a local PCA-based metric that measures the degree of semantic drift and provides the first quantitative framework to analyze how different aggregation mechanisms affect manifold structure.
  arXiv  Detail & Related papers  (2025-11-12T06:48:43Z)
• TerraFM: A Scalable Foundation Model for Unified Multisensor Earth Observation [65.74990259650984]
  We introduce TerraFM, a scalable self-supervised learning model that leverages globally distributed Sentinel-1 and Sentinel-2 imagery.<n>Our training strategy integrates local-global contrastive learning and introduces a dual-centering mechanism.<n>TerraFM achieves strong generalization on both classification and segmentation tasks, outperforming prior models on GEO-Bench and Copernicus-Bench.
  arXiv  Detail & Related papers  (2025-06-06T17:59:50Z)
• Spatial Knowledge Graph-Guided Multimodal Synthesis [78.11669780958657]
  We introduce a novel multimodal synthesis approach guided by spatial knowledge graphs, grounded in the concept of knowledge-to-data generation.<n>In experiments, data synthesized from diverse types of spatial knowledge, including direction and distance, enhance the spatial perception and reasoning abilities of MLLMs markedly.<n>We hope that the idea of knowledge-based data synthesis can advance the development of spatial intelligence.
  arXiv  Detail & Related papers  (2025-05-28T17:50:21Z)
• Self-Supervised Representation Learning for Geospatial Objects: A Survey [21.504978593542354]
  Self-supervised learning (SSL) has garnered increasing attention for its ability to learn effective and generalizable representations directly from data without extensive labeled supervision.<n>This paper presents a survey of SSL techniques specifically applied to or developed for geospatial objects in three primary geometric vector types: Point, Polyline, and Polygon.<n>We examine the emerging trends in SSL for geospatial objects, particularly the gradual advancements towards geospatial foundation models.
  arXiv  Detail & Related papers  (2024-08-22T05:28:22Z)
• A General Purpose Neural Architecture for Geospatial Systems [142.43454584836812]
  We present a roadmap towards the construction of a general-purpose neural architecture (GPNA) with a geospatial inductive bias. We envision how such a model may facilitate cooperation between members of the community.
  arXiv  Detail & Related papers  (2022-11-04T09:58:57Z)
• Scalable Self-Supervised Representation Learning from Spatiotemporal Motion Trajectories for Multimodal Computer Vision [0.0]
  We propose a self-supervised, unlabeled method for learning representations of geographic locations from GPS trajectories. We show that reachability embeddings are semantically meaningful representations and result in 4-23% gain in performance as measured using area under precision-recall curve (AUPRC) metric.
  arXiv  Detail & Related papers  (2022-10-07T02:41:02Z)
• Reachability Embeddings: Scalable Self-Supervised Representation Learning from Markovian Trajectories for Geospatial Computer Vision [0.0]
  We propose a self-supervised method for learning representations of geographic locations from unlabeled GPS trajectories. A scalable and distributed algorithm is presented to compute image-like representations, called reachability summaries. We show that reachability embeddings are semantically meaningful representations and result in 4-23% gain in performance.
  arXiv  Detail & Related papers  (2021-10-24T20:10:22Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.