Camera Calibration
Overview
Camera calibration estimates intrinsics such as focal length and distortion, and extrinsics such as camera pose relative to other cameras, robots, or sensors. Approaches range from target-based geometric optimization and bundle adjustment to target-free self-calibration, learned correspondence or parameter prediction, and refinements using motion, scene geometry, or external maps. Work spans single cameras, multi-camera rigs, and camera–LiDAR, radar, event-camera, and robot systems, with benchmarks testing reprojection, pose, reconstruction, and downstream task accuracy. A central tension is that low reprojection error does not always mean accurate 3D results, while robustness depends on informative observations, parameter observability, changing conditions, and reliable independent verification.
Latest papers
57 papers9 papers in the last four weeks, up 50% on the four weeks before. 0.1% of all new papers.