Enables high-fidelity 3D satellite surface reconstruction in a single forward pass without per-scene optimization.
March 20, 2026
Original Paper
SwiftGS: Episodic Priors for Immediate Satellite Surface Recovery
arXiv · 2603.18634
The Takeaway
Traditional 3D Gaussian Splatting requires expensive per-scene fitting; this model meta-learns episodic priors to achieve zero-shot reconstruction. This drastically reduces the compute time for large-scale geospatial monitoring from hours of optimization to milliseconds of inference.
From the abstract
Rapid, large-scale 3D reconstruction from multi-date satellite imagery is vital for environmental monitoring, urban planning, and disaster response, yet remains difficult due to illumination changes, sensor heterogeneity, and the cost of per-scene optimization. We introduce SwiftGS, a meta-learned system that reconstructs 3D surfaces in a single forward pass by predicting geometry-radiation-decoupled Gaussian primitives together with a lightweight SDF, replacing expensive per-scene fitting with