Search papers, labs, and topics across Lattice.
This paper addresses the challenges of calibrating multi-fisheye camera systems, revealing that calibration failures are primarily driven by poor intrinsic initialization rather than detector recall loss. By introducing CO-Calib, a framework that enhances calibration data construction through a robust learning-based target detector and an error-analysis-guided frame selector, the authors significantly improve calibration outcomes. Experimental results show an increase in the overall success rate from 68.1% to 99.3%, demonstrating enhanced extrinsic accuracy and stability in real-world applications.
Calibration failures in multi-fisheye systems are primarily due to intrinsic initialization issues, not just detector performance.
Reliable calibration of multi-fisheye camera systems remains challenging as rig size, camera arrangement diversity, and field of view increase. Existing pipelines can jointly optimize intrinsics, extrinsics, and target poses, but their success still depends heavily on empirical capture rules and the quality of the observations supplied to the solver. This paper studies this dependency through a failure-oriented analysis. We reveal that calibration failures are not sufficiently explained by detector recall loss or global image-plane distribution imbalance. Instead, the dominant failure factor lies in intrinsic initialization: observations with limited radial span couple focal scale with fisheye projection-shape parameters, producing ill-conditioned updates. Guided by this insight, we propose CO-Calib, a plug-in calibration-data construction framework that combines a robust learning-based target detector with an error-analysis-guided frame selector. CO-Calib constructs initialization-friendly anchors, co-visible multi-camera constraints, and coverage-completion frames without changing the existing calibration workflow or optimization backend. Extensive experiments on synthetic and real multi-fisheye systems demonstrate that CO-Calib improves the overall success rate from 68.1% to 99.3%, increases extrinsic accuracy, and augments real-world calibration stability. The source code will be made publicly available at https://github.com/HKUST-Aerial-Robotics/CO-Calib.
