SCAR: Satellite Imagery-Based Calibration for Aerial Recordings

TL;DR

SCAR tackles long-term calibration drift in aerial visual–inertial systems by leveraging georeferenced satellite imagery as a persistent global reference. It jointly refines camera intrinsics and camera–IMU extrinsics through a nonlinear factor-graph that fuses GNSS/INS pose priors with automatically generated image anchors from aerial–satellite matching and DEMs. The method demonstrates substantial reductions in reprojection error and marked improvements in visual localization accuracy across multiple campaigns and conditions, while being fully automatic and reusable through an open-source toolbox. These results show that long-term, scalable aerial calibration is achievable without manual interventions, enabling more robust fleet operations and retrospective recalibration from existing data. The work also documents limitations related to parallax, DEM quality, and near-nadir operation, outlining clear directions for extending SCAR to oblique views and additional sensing modalities.

Abstract

We introduce SCAR, a method for long-term auto-calibration refinement of aerial visual-inertial systems that exploits georeferenced satellite imagery as a persistent global reference. SCAR estimates both intrinsic and extrinsic parameters by aligning aerial images with 2D--3D correspondences derived from publicly available orthophotos and elevation models. In contrast to existing approaches that rely on dedicated calibration maneuvers or manually surveyed ground control points, our method leverages external geospatial data to detect and correct calibration degradation under field deployment conditions. We evaluate our approach on six large-scale aerial campaigns conducted over two years under diverse seasonal and environmental conditions. Across all sequences, SCAR consistently outperforms established baselines (Kalibr, COLMAP, VINS-Mono), reducing median reprojection error by a large margin, and translating these calibration gains into substantially lower visual localization rotation errors and higher pose accuracy. These results demonstrate that SCAR provides accurate, robust, and reproducible calibration over long-term aerial operations without the need for manual intervention.

Figures (6)

• Figure 1: Overview of the proposed SCAR framework for long-term INS–camera calibration refinement. (a) Aerial imagery is matched against satellite or orthophoto data to generate georeferenced anchors, while feature correspondences are tracked across frames to provide multi-view constraints for optimization. (b) Initial intrinsic and extrinsic calibrations are used to evaluate reprojection errors of these anchors. (c) Both absolute anchors and absolute pose measurements (GNSS/INS) are fused in a non-linear factor graph optimization, yielding refined intrinsic and extrinsic calibration parameters.
• Figure 2: Factor graph structure of SCAR. Camera poses $C_t$, anchors $X_j$, and intrinsics $K$ are modeled as variables, constrained by pose priors$\hat{P}^W_{\text{CAM},t}$, anchor priors$\hat{X}^W_j$, calibration prior$K_\text{init}$ and reprojection factors.
• Figure 3: Selected segments of both routes A (orange) and B (blue). For each route, segment 1 is used for calibration, segment 2 for validation.
• Figure 4: Reprojection examples showing reduced errors with SCAR (b,d) compared to Kalibr (a,c), with high (a) and low (c) initial error. The colormap in (a,b) ranges from 0 to 100 px and from 0 to 50 px in (c,d).
• Figure 5: Ablation study of SCAR. Shown is the change when removing nadir assumptions (“no nadir”), merging camera pose optimization into one step (“cam opt joint”), or skipping the final refinement (“no fine adjust”).