Research team of Professor Yoon Kuk-Jin developed a technique for joint estimation of camera orientation and vanishing points nonlinear Bayesian filtering
Research team of Professor Yoon Kuk-Jin developed a technique for joint estimation of camera orientation and vanishing points from an image sequence in a non-Manhattan world. A widely used approach for estimating camera orientation uses Manhattan world constraint which enables an estimation of the drift-free camera orientation, however this approach is neither effective because of noisy parallel line segments nor performable in non-Manhattan world scenes. In this work, the proposed method using nonlinear Bayesian filtering considerably improved the accuracy of camera orientation estimation and overcame the limited scene constraint by the virtue of the joint estimation framework. In addition, the outlier rejection technique based on the 2VP-3L method and the keyframe-based feature management technique enhanced the robustness to noisy and spurious parallel lines. Research team of Professor Yoon Kuk-Jin will be published in the International Journal of Computer Vision (IJCV) 2019 by Springer. This work was supported by Next-Generation Information Computing Development Program and Samsung Research Funding Center of Samsung Electronics.
Jeong-Kyun Lee, Kuk-Jin Yoon, Joint Estimation of Camera Orientation and Vanishing Points from an Image Sequence in a Non-Manhattan World, International Journal of Computer Vision