Noisy One-Point Homographies are Surprisingly Good
(2024) 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024 In Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition p.5125-5134- Abstract
Two-view homography estimation is a classic and fundamental problem in computer vision. While conceptually simple, the problem quickly becomes challenging when multiple planes are visible in the image pair. Even with correct matches, each individual plane (homography) might have a very low number of inliers when comparing to the set of all correspondences. In practice, this requires a large number of RANSAC iterations to generate a good model hypothesis. The current state-of-the-art methods therefore seek to reduce the sample size, from four point correspondences originally, by including additional information such as keypoint orientation/angles or local affine information. In this work, we continue in this direction and propose a novel... (More)
Two-view homography estimation is a classic and fundamental problem in computer vision. While conceptually simple, the problem quickly becomes challenging when multiple planes are visible in the image pair. Even with correct matches, each individual plane (homography) might have a very low number of inliers when comparing to the set of all correspondences. In practice, this requires a large number of RANSAC iterations to generate a good model hypothesis. The current state-of-the-art methods therefore seek to reduce the sample size, from four point correspondences originally, by including additional information such as keypoint orientation/angles or local affine information. In this work, we continue in this direction and propose a novel one-point solver that leverages different approximate constraints derived from the same auxiliary information. In experiments we obtain state-of-the-art results, with execution time speed-ups, on large benchmark datasets and show that it is more beneficial for the solver to be sample efficient compared to generating more accurate homographies.
(Less)
- author
- Ding, Yaqing
LU
; Astermark, Jonathan
LU
; Oskarsson, Magnus
LU
and Larsson, Viktor LU
- organization
- publishing date
- 2024
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- keywords
- Homography, Relative pose, SIFT
- host publication
- Proceedings - 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024
- series title
- Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
- pages
- 10 pages
- publisher
- IEEE Computer Society
- conference name
- 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024
- conference location
- Seattle, United States
- conference dates
- 2024-06-16 - 2024-06-22
- external identifiers
-
- scopus:85207303475
- ISSN
- 1063-6919
- ISBN
- 9798350353006
- DOI
- 10.1109/CVPR52733.2024.00490
- language
- English
- LU publication?
- yes
- additional info
- Publisher Copyright: © 2024 IEEE.
- id
- e49eff51-91f2-4a9f-b0ae-cae6e5d24733
- date added to LUP
- 2024-12-17 11:48:26
- date last changed
- 2025-04-04 15:07:11
@inproceedings{e49eff51-91f2-4a9f-b0ae-cae6e5d24733, abstract = {{<p>Two-view homography estimation is a classic and fundamental problem in computer vision. While conceptually simple, the problem quickly becomes challenging when multiple planes are visible in the image pair. Even with correct matches, each individual plane (homography) might have a very low number of inliers when comparing to the set of all correspondences. In practice, this requires a large number of RANSAC iterations to generate a good model hypothesis. The current state-of-the-art methods therefore seek to reduce the sample size, from four point correspondences originally, by including additional information such as keypoint orientation/angles or local affine information. In this work, we continue in this direction and propose a novel one-point solver that leverages different approximate constraints derived from the same auxiliary information. In experiments we obtain state-of-the-art results, with execution time speed-ups, on large benchmark datasets and show that it is more beneficial for the solver to be sample efficient compared to generating more accurate homographies.</p>}}, author = {{Ding, Yaqing and Astermark, Jonathan and Oskarsson, Magnus and Larsson, Viktor}}, booktitle = {{Proceedings - 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2024}}, isbn = {{9798350353006}}, issn = {{1063-6919}}, keywords = {{Homography; Relative pose; SIFT}}, language = {{eng}}, pages = {{5125--5134}}, publisher = {{IEEE Computer Society}}, series = {{Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition}}, title = {{Noisy One-Point Homographies are Surprisingly Good}}, url = {{http://dx.doi.org/10.1109/CVPR52733.2024.00490}}, doi = {{10.1109/CVPR52733.2024.00490}}, year = {{2024}}, }