AeroSCR: Recovering Stable Scene-Coordinate Decodability Under Scale and Rotation Shifts

Recent scene coordinate regression (SCR) methods have narrowed the robustness gap to structure-based localization through stronger local representations and improved generalization. Yet SCR remains vulnerable when query observations differ substantially from the mapping conditions. We investigate this gap in near-nadir aerial relocalization, where altitude and heading changes produce large shifts in image scale and in-plane orientation. Controlled experiments show that descriptor variation can preserve relative correspondence discrimination while destabilizing scene-coordinate predictions, revealing different robustness requirements for correspondence retrieval and coordinate decoding. Motivated by this distinction, AeroSCR expands decoder training coverage across scale changes, canonicalizes large heading changes in descriptor space, and learns tolerance to residual yaw errors. Direct metric supervision and retrieval-gated GeoCells further adapt SCR to weak aerial geometry and large-area maps. Across six scenes from AnyVisLoc, AeroSCR reaches 90.8% success at 5 m/5${}^{\circ }$, essentially matching HLoc at 90.2%, with 4.0$\times$ less scene-specific storage and 8.2$\times$ lower mean online latency. On the roughly 10 km${}^2$ Oslo-Large scene, it achieves 86.2% success with 6.0$\times$ less storage and 3.1$\times$ lower latency than HLoc.