Calibration-free structure-from-motion with calibrated radial trifocal tensors
(2020) 16th European Conference on Computer Vision, ECCV 2020 In Lecture Notes in Computer Science 12350. p.382-399- Abstract
- In this paper we consider the problem of Structure-from-Motion from images with unknown intrinsic calibration. Instead of estimating the internal camera parameters through some self-calibration procedure, we propose to use a subset of the reprojection constraints that is invariant to radial displacement. This allows us to recover metric 3D reconstructions without explicitly estimating the cameras’ focal length or radial distortion parameters. The weaker projection model makes initializing the reconstruction especially difficult. To handle this additional challenge we propose two novel minimal solvers for radial trifocal tensor estimation. We evaluate our approach on real images and show that even for extreme optical systems, such as... (More)
- In this paper we consider the problem of Structure-from-Motion from images with unknown intrinsic calibration. Instead of estimating the internal camera parameters through some self-calibration procedure, we propose to use a subset of the reprojection constraints that is invariant to radial displacement. This allows us to recover metric 3D reconstructions without explicitly estimating the cameras’ focal length or radial distortion parameters. The weaker projection model makes initializing the reconstruction especially difficult. To handle this additional challenge we propose two novel minimal solvers for radial trifocal tensor estimation. We evaluate our approach on real images and show that even for extreme optical systems, such as fisheye or catadioptric, we are able to get accurate reconstructions without performing any calibration. (Less)
Please use this url to cite or link to this publication:
https://lup.lub.lu.se/record/816a2bf9-58d8-4e07-80d4-9984a08d5c12
- author
- Larsson, Viktor LU ; Zobernig, Nicolas ; Taskin, Kasim and Pollefeys, Marc
- publishing date
- 2020
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- Computer Vision – ECCV 2020 : 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part V - 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part V
- series title
- Lecture Notes in Computer Science
- volume
- 12350
- pages
- 18 pages
- publisher
- Springer
- conference name
- 16th European Conference on Computer Vision, ECCV 2020
- conference location
- Glasgow, United Kingdom
- conference dates
- 2020-08-23 - 2020-08-28
- external identifiers
-
- scopus:85097421830
- ISSN
- 1611-3349
- 0302-9743
- ISBN
- 978-3-030-58451-1
- 978-3-030-58452-8
- DOI
- 10.1007/978-3-030-58558-7_23
- language
- English
- LU publication?
- no
- id
- 816a2bf9-58d8-4e07-80d4-9984a08d5c12
- date added to LUP
- 2022-09-06 13:17:14
- date last changed
- 2024-03-21 13:27:49
@inproceedings{816a2bf9-58d8-4e07-80d4-9984a08d5c12, abstract = {{In this paper we consider the problem of Structure-from-Motion from images with unknown intrinsic calibration. Instead of estimating the internal camera parameters through some self-calibration procedure, we propose to use a subset of the reprojection constraints that is invariant to radial displacement. This allows us to recover metric 3D reconstructions without explicitly estimating the cameras’ focal length or radial distortion parameters. The weaker projection model makes initializing the reconstruction especially difficult. To handle this additional challenge we propose two novel minimal solvers for radial trifocal tensor estimation. We evaluate our approach on real images and show that even for extreme optical systems, such as fisheye or catadioptric, we are able to get accurate reconstructions without performing any calibration.}}, author = {{Larsson, Viktor and Zobernig, Nicolas and Taskin, Kasim and Pollefeys, Marc}}, booktitle = {{Computer Vision – ECCV 2020 : 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part V}}, isbn = {{978-3-030-58451-1}}, issn = {{1611-3349}}, language = {{eng}}, pages = {{382--399}}, publisher = {{Springer}}, series = {{Lecture Notes in Computer Science}}, title = {{Calibration-free structure-from-motion with calibrated radial trifocal tensors}}, url = {{http://dx.doi.org/10.1007/978-3-030-58558-7_23}}, doi = {{10.1007/978-3-030-58558-7_23}}, volume = {{12350}}, year = {{2020}}, }