Search: onr:"swepub:oai:research.chalmers.se:c4fb2e1d-90a7-4c55-89c6-a27db6cbb6af" >
Back to the Feature...
Back to the Feature: Learning Robust Camera Localization from Pixels to Pose
-
- Sarlin, Paul-Edouard (author)
- Eidgenössische Technische Hochschule Zürich (ETH),Swiss Federal Institute of Technology in Zürich (ETH)
-
- Unagar, Ajaykumar (author)
- Eidgenössische Technische Hochschule Zürich (ETH),Swiss Federal Institute of Technology in Zürich (ETH)
-
Larsson, Mans (author)
-
show more...
-
- Germain, Hugo (author)
- École des Ponts ParisTech
-
- Toft, Carl, 1990 (author)
- Chalmers tekniska högskola,Chalmers University of Technology
-
- Larsson, Viktor (author)
- Eidgenössische Technische Hochschule Zürich (ETH),Swiss Federal Institute of Technology in Zürich (ETH)
-
- Pollefeys, Marc (author)
- Eidgenössische Technische Hochschule Zürich (ETH),Swiss Federal Institute of Technology in Zürich (ETH)
-
- Lepetit, Vincent (author)
- École des Ponts ParisTech
-
- Hammarstrand, Lars, 1979 (author)
- Chalmers tekniska högskola,Chalmers University of Technology
-
- Kahl, Fredrik, 1972 (author)
- Chalmers tekniska högskola,Chalmers University of Technology
-
- Sattler, Torsten, 1983 (author)
- Chalmers tekniska högskola,Chalmers University of Technology
-
show less...
-
(creator_code:org_t)
- 2021
- 2021
- English.
-
In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. - 1063-6919. ; , s. 3246-3256
- Related links:
-
https://research.cha...
-
show more...
-
https://doi.org/10.1...
-
show less...
Abstract
Subject headings
Close
- Camera pose estimation in known scenes is a 3D geometry task recently tackled by multiple learning algorithms. Many regress precise geometric quantities, like poses or 3D points, from an input image. This either fails to generalize to new viewpoints or ties the model parameters to a specific scene. In this paper, we go Back to the Feature: we argue that deep networks should focus on learning robust and invariant visual features, while the geometric estimation should be left to principled algorithms. We introduce PixLoc, a scene-agnostic neural network that estimates an accurate 6-DoF pose from an image and a 3D model. Our approach is based on the direct alignment of multiscale deep features, casting camera localization as metric learning. PixLoc learns strong data priors by end-to-end training from pixels to pose and exhibits exceptional generalization to new scenes by separating model parameters and scene geometry. The system can localize in large environments given coarse pose priors but also improve the accuracy of sparse feature matching by jointly refining keypoints and poses with little overhead. The code will be publicly available at github.com/cvg/pixloc.
Subject headings
- NATURVETENSKAP -- Data- och informationsvetenskap -- Bioinformatik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Bioinformatics (hsv//eng)
- TEKNIK OCH TEKNOLOGIER -- Annan teknik -- Mediateknik (hsv//swe)
- ENGINEERING AND TECHNOLOGY -- Other Engineering and Technologies -- Media Engineering (hsv//eng)
- NATURVETENSKAP -- Data- och informationsvetenskap -- Datorseende och robotik (hsv//swe)
- NATURAL SCIENCES -- Computer and Information Sciences -- Computer Vision and Robotics (hsv//eng)
Publication and Content Type
- kon (subject category)
- ref (subject category)
Find in a library
To the university's database
- By the author/editor
-
Sarlin, Paul-Edo ...
-
Unagar, Ajaykuma ...
-
Larsson, Mans
-
Germain, Hugo
-
Toft, Carl, 1990
-
Larsson, Viktor
-
show more...
-
Pollefeys, Marc
-
Lepetit, Vincent
-
Hammarstrand, La ...
-
Kahl, Fredrik, 1 ...
-
Sattler, Torsten ...
-
show less...
- About the subject
-
- NATURAL SCIENCES
-
NATURAL SCIENCES
-
and Computer and Inf ...
-
and Bioinformatics
-
- ENGINEERING AND TECHNOLOGY
-
ENGINEERING AND ...
-
and Other Engineerin ...
-
and Media Engineerin ...
-
- NATURAL SCIENCES
-
NATURAL SCIENCES
-
and Computer and Inf ...
-
and Computer Vision ...
- Articles in the publication
-
Proceedings of t ...
- By the university
-
Chalmers University of Technology