arXiv:2608. 20149v1 Announce Type: new Abstract: This article explores the performance of analytical and neural-based hillshading methods in a dense urban environment using high-resolution digital elevation model (DEM) and digital surface model (DSM) data for downtown Calgary.
arXiv:2608. 20026v1 Announce Type: cross Abstract: Streetscape quality has become a central concern in contemporary urban planning, particularly within the framework of the pedestrian-friendly 15-minute city, where walkability and public-space quality are increasingly recognized as key determinants of urban performance.
arXiv:2606. 00404v1 Announce Type: cross Abstract: Implicit neural representations (INRs) model a signal as a continuous coordinate-to-value function.
The study evaluates how much street‑view imagery contributes to urban attribute prediction beyond existing public data. By comparing image‑based models with seven attributes from five public sources and three vision‑language models, the authors find that images outperform other data for building type, function, and low‑rise floor count, while existing data match or exceed image performance for road damage, curb ramps, and house price. The benefit of images varies with visual legibility and local data coverage, suggesting that image value depends on how well the scene is captured and how much complementary data is available.
The paper presents a method for enhancing coarse 5 m digital surface models (DSMs) to 0.5 m resolution by guiding a denoising diffusion process with high‑resolution spectral images. This approach transfers fine visual details—such as crisp outlines and roof structures—from the imagery into the elevation maps, yielding more accurate surface geometry than traditional interpolation or filtering. Experiments on Central European cities show that the resulting DSMs exhibit improved structural detail and overall quality.
arXiv:2607. 22342v1 Announce Type: new Abstract: The development of effective urban climate adaptation strategies requires comprehensive spatial information on rooftops and buildings, since such information underpins the assessment of ecosystem services provided by green infrastructure, particularly for urban heat island (UHI) mitigation.
arXiv:2607. 14756v1 Announce Type: new Abstract: This research investigates the potential of Vision-Language Models (VLMs) to infer building typologies: Construction, Current Use, and Storeys from Google Street View (GSV) images.
arXiv:2606. 15890v1 Announce Type: new Abstract: Understanding urban wellbeing from multimodal data requires integrating heterogeneous spatial and temporal signals, posing significant challenges for current multimodal large language models (MLLMs).
The paper proposes a method to enhance coarse 5 m digital surface models (DSMs) to 0.5 m resolution by guiding the super‑resolution process with high‑resolution spectral images. It uses denoising diffusion to transfer image‑visible details, such as crisp outlines and roof structures, into the elevation maps, achieving more accurate surface geometry than traditional interpolation or filtering. Experiments on Central European cities show that the approach yields high‑quality DSMs with improved structural detail.
The paper proposes a method to adapt the Depth Anything V2 (DAV2) zero‑shot relative depth model for estimating lunar surface height. By fine‑tuning DAV2 with publicly available stereophotogrammetry‑derived DEM data, the authors achieve a significant performance boost over the unadapted zero‑shot model. This improved estimator can provide more accurate relative height information useful for hazard detection in future ESA lunar landings.
BEACON is a tri‑modal contrastive learning framework that enriches AlphaEarth embeddings by aligning physical representations from Earth‑observation imagery with semantic POI text and human behavioral POI visitation data, while keeping the deployed model image‑only. In a Houston case study, BEACON outperformed six baselines on nine downstream tasks, achieving up to 43% higher R² for obesity prevalence, 34% for poor mental health, and 22% for median household income under a linear probe.
ImplicitTerrainV2 introduces a wavelet-guided, spatially adaptive neural representation for digital elevation models (DEMs). It uses a wavelet complexity field to localize high-frequency capacity to complex terrain, adaptive sampling to focus training, and gradient matching to preserve smooth manifold structure. After mixed-precision quantization and entropy coding, the model achieves 1.23 bpp with only a 0.28 dB PSNR loss, outperforming prior work by 5.70 dB while using 3.2× fewer parameters and training in 55 s per tile on a single GPU.
The paper introduces a deep‑learning framework that estimates building age using widely available multi‑source geospatial data, including satellite imagery, Digital Surface Models, and building footprints. It employs a pretrained backbone with fine‑tuning adapters, raster‑derived and property‑derived fusion modules, and a pyramid decoder with Cosine‑guided Mixture‑of‑Experts for pixel‑level predictions, while instance‑level predictions use attention pooling and Property‑Conditional Adaptive Instance Normalization. A Temporal Ordinal Constraint Loss enforces the natural ordering of building ages, and the authors provide a new Hong Kong building age benchmark along with open‑source code and datasets.