We compute camera pose parameters from a sequence of images using a sequential estimation procedure. Taking into account both the camera motion model and the noisy observation model, the results tend to be more accurate and more robust compared to when the observations are considered alone. To combine the camera pose history and the uncertainty of image point measurements, the Extended Kalman Filter (EKF) is employed.
Mehralian, Mohammad Amin; Soryani, Mohsen: 'EKFPnP: extended Kalman filter for camera pose estimation in a sequence of images', IET Image Processing, 2020, 14, (15), p. 3774-3780, DOI: 10.1049/iet-ipr.2020.0606
IET Digital Library: [https://digital-library.theiet.org/content/journals/10.1049/iet-ipr.2020.0606]
PDF: [https://ietresearch.onlinelibrary.wiley.com/doi/epdf/10.1049/iet-ipr.2020.0606]
This toolbox is an extension of the toolbox provided by the authors of MLPnP, CEPPnP and OPnP We extended it to show the use of EKFPnP.
main_ordinary_3d.m
main_planar.m
main_ordinary_3d_sigma.m
main_planar_sigma.m
main_ordinary_3d_time.m
main_planar_time.m
main_box.m
main_sfm.m
This toolbox is based on the toolbox provided in:
MLPnP: Urban, S.; Leitloff, J.; Hinz, S.. MLPNP - A REAL-TIME MAXIMUM LIKELIHOOD SOLUTION TO THE PERSPECTIVE-N-POINT PROBLEM. ISPRS Annals of Photogrammetry, https://github.com/urbste/MLPnP_matlab
CEPPnP: Luis Ferraz, Xavier Binefa, Francesc Moreno-Noguer. Leveraging Feature Uncertainty in the PnP Problem. In Proceedings of BMVC, 2014.
REPPnP: Luis Ferraz, Xavier Binefa, Francesc Moreno-Noguer. Very Fast Solution to the PnP Problem with Algebraic Outlier Rejection. In Proceedings of CVPR, 2014.
OPnP: Y. Zheng, Y. Kuang, S. Sugimoto, K. Astro ?m, and M. Okutomi. Revisiting the pnp problem: A fast, general and opti- mal solution. In ICCV, pages 4321?4328, 2013.
ASPnP: Y. Zheng, S. Sugimoto, and M. Okutomi. Aspnp: An accurate and scalable solution to the perspective-n-point problem. Trans. on Information and Systems, 96(7):1525?1535, 2013.