Scipediacontent: Scipediacontent moved page Draft Content 178430458 to Pajdla et al 2020a

2021-02-01T20:08:01Z

Scipediacontent moved page Draft Content 178430458 to Pajdla et al 2020a

Scipediacontent: Created page with " == Abstract == The quality and speed of Structure from Motion (SfM) methods depend significantly on the camera model chosen for the reconstruction. In most of the SfM pipeli..."

2021-02-01T20:07:53Z

Created page with " == Abstract == The quality and speed of Structure from Motion (SfM) methods depend significantly on the camera model chosen for the reconstruction. In most of the SfM pipeli..."

New page

== Abstract ==

The quality and speed of Structure from Motion (SfM) methods depend significantly on the camera model chosen for the reconstruction. In most of the SfM pipelines, the camera model is manually chosen by the user. In this paper, we present a new automatic method for camera model selection in large scale SfM that is based on efficient uncertainty evaluation. We first perform an extensive comparison of classical model selection based on known Information Criteria and show that they do not provide sufficiently accurate results when applied to camera model selection. Then we propose a new Accuracy-based Criterion, which evaluates an efficient approximation of the uncertainty of the estimated parameters in tested models. Using the new criterion, we design a camera model selection method and fine-tune it by machine learning. Our simulated and real experiments demonstrate a significant increase in reconstruction quality as well as a considerable speedup of the SfM process.

Document type: Conference object

== Full document ==
<pdf>Media:Draft_Content_178430458-beopen2548-7000-document.pdf</pdf>

== Original document ==

The different versions of the original document can be found in:

* [http://dx.doi.org/10.5281/zenodo.4024106 http://dx.doi.org/10.5281/zenodo.4024106] under the license https://creativecommons.org/licenses/by

* [http://dx.doi.org/10.5281/zenodo.4024107 http://dx.doi.org/10.5281/zenodo.4024107] under the license https://creativecommons.org/licenses/by/4.0/legalcode

* [https://zenodo.org/record/4024107 https://zenodo.org/record/4024107] under the license https://creativecommons.org/licenses/by/4.0/legalcode

* [https://zenodo.org/record/4024107/files/Uncertainty_Based_Camera_Model_Selection.pdf https://zenodo.org/record/4024107/files/Uncertainty_Based_Camera_Model_Selection.pdf]

* [http://xplorestaging.ieee.org/ielx7/9142308/9156271/09156313.pdf?arnumber=9156313 http://xplorestaging.ieee.org/ielx7/9142308/9156271/09156313.pdf?arnumber=9156313],
: [http://dx.doi.org/10.1109/cvpr42600.2020.00603 http://dx.doi.org/10.1109/cvpr42600.2020.00603] under the license cc-by

* [https://dblp.uni-trier.de/db/conf/cvpr/cvpr2020.html#PolicSAKP20 https://dblp.uni-trier.de/db/conf/cvpr/cvpr2020.html#PolicSAKP20],
: [http://openaccess.thecvf.com/content_CVPR_2020/papers/Polic_Uncertainty_Based_Camera_Model_Selection_CVPR_2020_paper.pdf http://openaccess.thecvf.com/content_CVPR_2020/papers/Polic_Uncertainty_Based_Camera_Model_Selection_CVPR_2020_paper.pdf],
: [https://ieeexplore.ieee.org/document/9156313 https://ieeexplore.ieee.org/document/9156313],
: [http://openaccess.thecvf.com/content_CVPR_2020/html/Polic_Uncertainty_Based_Camera_Model_Selection_CVPR_2020_paper.html http://openaccess.thecvf.com/content_CVPR_2020/html/Polic_Uncertainty_Based_Camera_Model_Selection_CVPR_2020_paper.html],
: [https://academic.microsoft.com/#/detail/3034634220 https://academic.microsoft.com/#/detail/3034634220]

DOIS: 10.5281/zenodo.4024106 10.1109/cvpr42600.2020.00603 10.5281/zenodo.4024107

← Older revision	Revision as of 20:08, 1 February 2021
(No difference)

Pajdla et al 2020a - Revision history

Scipediacontent: Scipediacontent moved page Draft Content 178430458 to Pajdla et al 2020a

Scipediacontent: Created page with " == Abstract == The quality and speed of Structure from Motion (SfM) methods depend significantly on the camera model chosen for the reconstruction. In most of the SfM pipeli..."