| HAL-Inria | Publications, software ... of Inria's scientists |
Conference papers
Are Large-Scale 3D Models Really Necessary for Accurate Visual Localization?
Abstract : Accurate visual localization is a key technology for autonomous navigation. 3D structure-based methods employ 3D models of the scene to estimate the full 6DOF pose of a camera very accurately. However, constructing (and extending) large-scale 3D models is still a significant challenge. In contrast, 2D image retrieval-based methods only require a database of geo-tagged images, which is trivial to construct and to maintain. They are often considered inaccurate since they only approximate the positions of the cameras. Yet, the exact camera pose can theoretically be recovered when enough relevant database images are retrieved. In this paper, we demonstrate experimentally that large-scale 3D models are not strictly necessary for accurate visual localization. We create reference poses for a large and challenging urban dataset. Using these poses, we show that combining image-based methods with local reconstructions results in a pose accuracy similar to the state-of-the-art structure-based methods. Our results suggest that we might want to reconsider the current approach for accurate large-scale localization.
Cited literature [56 references]
https://hal.inria.fr/hal-01513083
Contributor : Josef Sivic <>
Submitted on : Monday, April 24, 2017 - 3:59:44 PM
Last modification on : Thursday, May 27, 2021 - 1:54:06 PM
Long-term archiving on: : Tuesday, July 25, 2017 - 4:54:00 PM
Contributor : Josef Sivic <>
Submitted on : Monday, April 24, 2017 - 3:59:44 PM
Last modification on : Thursday, May 27, 2021 - 1:54:06 PM
Long-term archiving on: : Tuesday, July 25, 2017 - 4:54:00 PM
File
Sattler17.pdf
Files produced by the author(s)