Papers for
geospatial data providers
Papers whose findings have a practical use for this group, as judged from the abstract. Open a paper to read what it means in practice.
RSLiveBench UAV benchmark tracks aerial spatial intelligence progress
RSLiveBench-UAV: A Continually Evolving Vision-Language Benchmark for Aerial Spatial Intelligence
Abstract: Aerial spatial intelligence (ASI) requires models to understand objects and spatial relationships across changing viewpoints and scales. Evaluating these abilities as models evolve calls for fresh physical observations, yet producing new aerial spatial questions requires repeatedly establishing entity identities, geometric references, and cross-view correspondences. We introduce \textbf{RSLiveBench-UAV}, a continually evolving vision-language benchmark driven by recurring acquisition of real UAV imagery. Its update protocol admits images captured after the preceding release and delays answer disclosure. An entity-centered production workflow uses geometry-assisted association to link repeated observations of physical vehicles, reuse human-verified attributes, and combine them with view-specific annotations and spatial references for traceable question generation. All tasks use deterministic scoring without model judges. We evaluate 30 general-purpose, remote-sensing, and embodied models on 35,491 question records from ten scenes, covering 17 tasks across 2D perception, 3D perception, cross-view perception, and reasoning. All 18 tested embodied models outperform the three selected remote-sensing comparators in Overall, although the strongest embodied model scores 43.29 versus 59.88 for GPT-6-Astra. Fine-grained vehicle recognition remains challenging, and performance varies across heights and camera pitches. Retrospective scene-group comparisons and rolling-window replay characterize score-pattern consistency under scene renewal. RSLiveBench-UAV connects fresh physical observations, entity-centered answer production, and systematic model evaluation to support continual assessment of ASI. Benchmark will release at https://rslivebench.space
Vector tile size affects map data compression efficiency
Does Tile Granularity Matter? An Empirical Study of Vector Tile Compression for Map Rendering with MapLibre Tiles
Abstract: Vector tiles are delivered as pyramids of small, independently encoded units, whose spatial granularity is largely determined by the tiling scheme. We investigate how this granularity affects compression in column-oriented MapLibre Tiles (MLT), using feature-oriented Mapbox Vector Tiles (MVT) as a baseline. Five heterogeneous geospatial datasets were encoded as conventional tile pyramids and as two denser variants grouping the area of 2 x 2 or 4 x 4 conventional tiles into one encoding unit at approximately matched logical levels of detail. Compared with conventional tiling, the 4 x 4 variant increased MLT's raw size saving over MVT by 1.0-7.3 percentage points across all five datasets. After per-tile gzip, the saving increased by 1.5-6.4 points in four datasets and decreased by 1.4 points in one. Coarser granularity lengthened MLT streams and reduced structural overhead in every dataset. The largest raw gains occurred where conventional tiles had the shortest streams: as mean values per stream rose from 27 to 2,075 across datasets, the raw gain decreased from 7.3 to 1.0 points. Encoded payload bytes per value also declined, with the strongest reductions in two datasets. For the two sparse datasets, coarser units reduced bytes per vertex across both formats, with or without gzip; the resulting 4 x 4 MLT+gzip archives were 71% and 33% smaller than the conventional ones. Thus, the measured compression advantage of MLT depends on spatial encoding-unit granularity, and conventional fine-grained tiling can understate the potential of column-oriented compression.
Crowdsourced disaster photos geolocated using vision and mapping AI
A Decade of PhotoMappers: A Longitudinal Study of Crowdsourced Disaster Photos Geolocalization with Vision-Language Models and Geospatial Reasoning
Abstract: Crowdsourced disaster imagery provides timely, ground-level observations when conventional disaster mapping is delayed or incomplete, yet its geographic value depends on where observations are produced, how they are interpreted, and whether they can be reliably located as actionable Geospatial Intelligence (GEOINT). Using more than a decade of PhotoMappers, a crowdsourced disaster-mapping initiative, this study examines the changing geography of disaster information and asks how emerging GeoAI models compare with human geographic interpretation. We combine longitudinal spatial analysis with cross-view geolocalization using street view imagery (SVI) and remote sensing imagery (RSI), and evaluate vision-language models (VLMs) reasoning through 30 indicators spanning geographic cues and reasoning evidence. The results reveal a geographically and temporally uneven archive shaped by changing disaster types, Community Lifelines, and operational priorities. Integrating VGI, SVI, and RSI with Sinkhorn re-ranking achieves 40.90% R@1, 61.10% R@5, and 69.45% R@10, while yearly R@1 varies markedly from 25.00% to 65.07%. Geographic reasoning is similarly uneven: Qwen3.6-27B achieves an overall agreement of 67% compared with 51% for InternVL3-8B, with both models showing greater difficulty with localized and place-specific evidence than broad spatial configurations. We argue that the value of GeoAI lies less in replacing volunteered geographic interpretation than in complementing it, scaling candidate retrieval while retaining human judgment for ambiguous and consequential cases. This provides a pathway toward more human-centered and responsible GeoAI for disaster mapping. All data and code are openly available at https://github.com/bobleegogogo/PhotoMappers-VLM.
GeoCR removes clouds from satellite images across sensors and bands
GeoCR: Learning a Generalist Cloud Removal Prior from Heterogeneous Observations
Abstract: Cloud removal methods are typically specialized to individual datasets and input configurations, limiting reuse across sensors, spectral bands, and observation settings. We introduce GeoCR, a generalist model that unifies RGB-only-based CR and multispectral-based CR from single- or multi-temporal cloudy observations, with optional SAR guidance, within a single network. To accommodate different spectral and sensing domains, compact input and output stems extend a pretrained RGB autoencoder while keeping its encoder and decoder trunks frozen. This shared latent interface enables a single flow transformer to jointly model clean RGB and non-RGB latents, conditioned on separate cloudy-observation streams and optional SAR tokens. Through joint pretraining on the training splits of ten datasets comprising 883,331 cloud-free target images, GeoCR learns a shared cloud removal prior across these heterogeneous configurations. The same pretrained checkpoint supports direct inference without dataset-specific fine-tuning and efficient adaptation through low-rank adaptation (LoRA). We evaluate GeoCR against general image restoration and cloud removal methods on test splits of the contributing datasets under full-band and RGB-only settings. GeoCR achieves the best FID and DISTS on full-band SEN12MS-CR and Sen2_MTC_New and RGB-only CUHK-CR2, outperforming existing models and demonstrating the effectiveness of a reusable generative model across diverse settings.
Diffusion models improve 3D city maps from satellite images
Enhancing Photogrammetric Digital Surface Models with Pretrained Diffusion Models and Multimodal Conditioning
Abstract: Large-scale Digital Surface Models (DSMs) can be produced cost-effectively from satellite images via stereo-photogrammetry. However, the resulting 3D maps are often contaminated by noise, outliers, and voids. On the other hand, aerial LiDAR provides high-accuracy elevation measurements at a substantially higher cost. In this work, we study diffusion models conditioned both on photogrammetric DSMs and Pléiades imagery to refine vertically co-registered DSMs. We introduce a modified Stable Diffusion 3 architecture with a pruned text stream and a patch-wise normalization strategy, enabling stable training on LiDAR data and transfer from natural images to elevation maps. Experiments in French cities demonstrate that multimodal conditioning improves elevation accuracy, reducing Dense Urban RMSE from 6.00 to 3.45 m in the in-context cities and from 4.16 to 2.77 m in the held-out city of Bordeaux.