Agentic Building-Aware Satellite Gaussian Splatting for Auditable Urban DSM Reconstruction

A building-aware satellite Gaussian Splatting workflow using Segment Anything-derived building masks and an Agentic Reconstruction Controller reduced building-region DSM MAE from 0.844 m to 0.806 m on the DFC2019 JAX_004 scene, according to arXiv paper 2609.25578v1. A staged schedule improved full-scene MAE from 1.362 m to 1.349 m while retaining the building gain, and across four JAX scenes the Agent selected validated policies for general and building-focused DSM objectives while producing building-inventory metadata and per-scene decision records. The system combines semantic priors, policy selection, region-specific DSM metrics, and DSM-derived GIS surface products for auditable urban 3D analysis.

arXiv:2609.25578v1 Announce Type: new Abstract: Urban-scale 3D reconstruction from satellite imagery supports disaster response, city monitoring, and geospatial digital twins, yet neural rendering methods typically optimize average visual fidelity rather than the structures that analysts inspect first: buildings. We present an agentic building-aware satellite Gaussian Splatting workflow that uses Segment Anything-derived building masks as semantic priors and an Agentic Reconstruction Controller to select, verify, and record DSM reconstruction policies. On the DFC2019 JAX\ 004 scene, building-aware weighting reduces building-region DSM MAE from 0.844 m to 0.806 m, showing that semantic priors can shift reconstruction capacity toward analyst-critical regions. A staged schedule provides a balanced operating point, improving full-scene MAE from 1.362 m to 1.349 m while retaining a building gain. Across four JAX scenes, the Agent selects validated policies for both general DSM and building-focused DSM objectives, and produces building-inventory metadata and per-scene decision records. The system combines semantic priors, policy selection, region-specific DSM metrics, and DSM-derived GIS surface products for auditable urban 3D analysis.