This is the authors' abstract. We don't add key points for this paper.
Accurate and spatially consistent digital representations of the built environment are essential for urban infrastructure analysis, sustainability assessment, and planning, particularly in countries experiencing rapid urbanisation, such as India. GlobalBuildingAtlas (GBA) is a recently released open-access global dataset that provides individual building footprints, building heights, and level of detail-1 (LoD1) three-dimensional building models. However, its suitability for digital urban infrastructure applications across diverse urban contexts remains largely unevaluated. This study presents an independent assessment of GBA at the individual-building scale, focusing on footprint completeness and building height accuracy using independent reference data. Building footprint completeness was evaluated across five gated residential communities using reference building counts derived from automated extraction applied to very high-resolution satellite imagery, while building height accuracy was assessed for representative structures intersected by ICESat-2 laser altimetry ground tracks using photon-based height estimates. Results show that GBA achieves high building footprint completeness in planned low-rise residential developments, with completeness ratios exceeding 99% and minimal count discrepancies, supporting its use as a reliable digital representation of horizontal urban infrastructure. In contrast, building height estimates exhibit a systematic negative bias, with errors increasing for mid-rise and high-rise buildings, consistent with limitations associated with monocular optical imagery–based height inference, the uneven geographic distribution of LiDAR training data, and the geometric abstraction inherent to LoD1 building representations. Overall, the findings indicate that GBA provides a robust foundation for digital urban infrastructure and sustainability-oriented applications in India, while underscoring the need for caution and region-specific validation when its height component is applied in vertically complex urban environments.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Dandabathula et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: