Papers by Zuzana Kukelova
arXiv (Cornell University), Mar 28, 2023
Bookmarks Related papers MentionsView impact
arXiv (Cornell University), Jan 16, 2023
Bookmarks Related papers MentionsView impact
2023 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)
Bookmarks Related papers MentionsView impact
ArXiv, 2022
This paper proposes the geometric relationship of epipolar geometry and orientation- and scale-co... more This paper proposes the geometric relationship of epipolar geometry and orientation- and scale-covariant, e.g., SIFT, features. We derive a new linear constraint relating the unknown elements of the fundamental matrix and the orientation and scale. This equation can be used together with the well-known epipolar constraint to, e.g., estimate the fundamental matrix from four SIFT correspondences, essential matrix from three, and to solve the semi-calibrated case from three correspondences. Requiring fewer correspondences than the well-known point-based approaches (e.g., 5PT, 6PT and 7PT solvers) for epipolar geometry estimation makes RANSAC-like randomized robust estimation significantly faster. The proposed constraint is tested on a number of problems in a synthetic environment and on publicly available real-world datasets on more than 80000 image pairs. It is superior to the state-of-the-art in terms of processing time while often leading to more accurate results.
Bookmarks Related papers MentionsView impact
Computer Vision – ACCV 2018, 2019
Bookmarks Related papers MentionsView impact
2021 IEEE/CVF International Conference on Computer Vision (ICCV), 2021
Bookmarks Related papers MentionsView impact
2021 International Conference on 3D Vision (3DV), 2021
We consider the problem of stitching image sequences with cameras undergoing pure rotational moti... more We consider the problem of stitching image sequences with cameras undergoing pure rotational motion. We leverage the assumption of a locally constant rotation axis, i.e., neighboring frames have a shared but unknown rotation axis. This assumption holds in many common image capturing scenarios, e.g., panoramic sweeping motions. Using this additional constraint, we develop techniques for three-view camera rotation estimation; a minimal solver for the two-view estimation with a known rotation axis; and a globally optimal robust estimator for the two-view case. We show on publicly available datasets that the proposed methods lead to camera rotation estimation superior to the state-of-the-art in terms of accuracy with comparable run-time. The source code will be made available.
Bookmarks Related papers MentionsView impact
2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019
Bookmarks Related papers MentionsView impact
2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021
Bookmarks Related papers MentionsView impact
Computer Vision – ECCV 2020, 2020
Bookmarks Related papers MentionsView impact
2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020
Bookmarks Related papers MentionsView impact
2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018
Bookmarks Related papers MentionsView impact
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020
Bookmarks Related papers MentionsView impact
Lecture Notes in Computer Science, 2010
Bookmarks Related papers MentionsView impact
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011
Bookmarks Related papers MentionsView impact
Computer Vision and Image Understanding, 2010
Bookmarks Related papers MentionsView impact
2019 IEEE/CVF International Conference on Computer Vision (ICCV), 2019
Bookmarks Related papers MentionsView impact
The internal geometry of most modern consumer cameras is not adequately described by the perspect... more The internal geometry of most modern consumer cameras is not adequately described by the perspective projection. Almost all cameras exhibit some radial lens distortion and are equipped with an electronic rolling shutter that induces distortions when the camera moves during the image capture. When focal length has not been calibrated offline, the parameters that describe the radial and rolling shutter distortions are usually unknown. While for global shutter cameras, minimal solvers for the absolute camera pose and unknown focal length and radial distortion are available, solvers for the rolling shutter were missing. We present the first minimal solutions for the absolute pose of a rolling shutter camera with unknown rolling shutter parameters, focal length, and radial distortion. Our new minimal solvers combine iterative schemes designed for calibrated rolling shutter cameras with fast generalized eigenvalue and Groebner basis solvers. In a series of experiments, with both synthetic...
Bookmarks Related papers MentionsView impact
Computer Vision – ACCV 2020
Bookmarks Related papers MentionsView impact
Computer Vision – ECCV 2020
Bookmarks Related papers MentionsView impact
Uploads
Papers by Zuzana Kukelova