Songyou Peng and Peter Sturm, “Calibration Wizard: A Guidance System for Camera Calibration Based on Modelling Geometric and Corner Uncertainty,” in 2019 IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, Korea (South), Oct. 2019, pp. 1497–1505, doi: 10.1109/ICCV.2019.00158 [Online]. Available at: https://ieeexplore.ieee.org/document/9009540/. [Accessed: October 20, 2024]
@inproceedings{peng_calibration_2019,
address = {Seoul, Korea (South)},
title = {Calibration {Wizard}: {A} {Guidance} {System} for {Camera} {Calibration} {Based} on {Modelling} {Geometric} and {Corner} {Uncertainty}},
copyright = {https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html},
isbn = {978-1-72814-803-8},
shorttitle = {Calibration {Wizard}},
url = {https://ieeexplore.ieee.org/document/9009540/},
doi = {10.1109/ICCV.2019.00158},
language = {en},
urldate = {2024-10-20},
booktitle = {2019 {IEEE}/{CVF} {International} {Conference} on {Computer} {Vision} ({ICCV})},
publisher = {IEEE},
author = {Peng, Songyou and Sturm, Peter},
month = oct,
year = {2019},
pages = {1497--1505},
freepdf = {https://openaccess.thecvf.com/content_ICCV_2019/papers/Peng_Calibration_Wizard_A_Guidance_System_for_Camera_Calibration_Based_on_ICCV_2019_paper.pdf},
tldr = {Uses three freely-acquired poses to initialize, creates an optimization problem for the next pose such that the expected uncertainty of the intrinsic camera parameters is minimized. The process is to formulate the calibration problem as geometric reprojection error, and Jacobian matrices are computed. The data is extended to a hypothetical next pose, the next pose and intrinsic parameters are parameterized within the Jacobian. Through some matrix transformations, the covariance matrix of the intrinsic parameters can be extracted using the Jacobian. Corner uncertainty is incorporated, as poses that reduce uncertainty may be perpendicular to the image plane and be unusable. Code is available but in Matlab.},
code = {https://github.com/pengsongyou/CalibrationWizard}
}
It is well known that the accuracy of a calibration depends strongly on the choice of camera poses from which images of a calibration object are acquired. We present a system – Calibration Wizard – that interactively guides a user towards taking optimal calibration images. For each new image to be taken, the system computes, from all previously acquired images, the pose that leads to the globally maximum reduction of expected uncertainty on intrinsic parameters and then guides the user towards that pose. We also show how to incorporate uncertainty in corner point position in a novel principled manner, for both, calibration and computation of the next best pose. Synthetic and realworld experiments are performed to demonstrate the effectiveness of Calibration Wizard.
tl;dr: Uses three freely-acquired poses to initialize, creates an optimization problem for the next pose such that the expected uncertainty of the intrinsic camera parameters is minimized. The process is to formulate the calibration problem as geometric reprojection error, and Jacobian matrices are computed. The data is extended to a hypothetical next pose, the next pose and intrinsic parameters are parameterized within the Jacobian. Through some matrix transformations, the covariance matrix of the intrinsic parameters can be extracted using the Jacobian. Corner uncertainty is incorporated, as poses that reduce uncertainty may be perpendicular to the image plane and be unusable. Code is available but in Matlab.
Calibration Wizard: A Guidance System for Camera Calibration Based on Modelling Geometric and Corner Uncertainty.
Publisher: http://doi.org/10.1109/ICCV.2019.00158
h/t Amy Tabbpdf: https://openaccess.thecvf.com/content_ICCV_2019/papers/Peng_Calibration_Wizard_A_Guidance_System_for_Camera_Calibration_Based_on_ICCV_2019_paper.pdf
code: https://github.com/pengsongyou/CalibrationWizard