AnchorHEC: Targetless Hand-Eye Calibration of Wrist and Fixed Cameras
AnchorHEC: Targetless Hand-Eye Calibration of Wrist and Fixed Cameras
This paper introduces AnchorHEC, a targetless hand-eye calibration framework for wrist and fixed cameras that uses RGB images, forward kinematics, and a shared 3D foundation model. The method initializes both camera types using FK and a shared 3D reconstruction, applies FK-constrained bundle adjustment to reduce pose and scale errors, and relaxes constraints if refinement fails or inliers are insufficient. It then triangulates metric points to estimate fixed-camera poses. Across three backbones, simulation calibration errors decrease by 86.4% in translation and 71.6% in rotation relative to reconstruction-based baselines, with validation performed on a physical robot.
The authors present AnchorHEC, a targetless hand-eye calibration framework for wrist and fixed cameras that uses RGB images, forward kinematics, and a shared 3D foundation model; simulation results report up to 86.4% reduction in translation error and 71.6% in rotation error relative to reconstruction-based baselines, with validation on a physical robot.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
use the refined wrist transform and FK to triangulate metric points and estimate fixed-camera poses. Across three backbones, calibration errors in simulation decrease by 86.4% in translation and 71.6% in rotation, averaged over both camera types, relative to reconstruction-based baselines. We also validate AnchorHEC on a physical robot. Project page: https://hwanggb.github.io/AnchorHEC/
Open source S4
Source：arXiv Robotics — research abstracts · arxiv.org