Accurate Localization and Pose Estimation for Large 3D Models

Linus Svärm, Olof Enqvist, Magnus Oskarsson, Fredrik Kahl

Research output: Chapter in Book/Report/Conference proceedingPaper in conference proceedingpeer-review

Abstract

We consider the problem of localizing a novel image in a large 3D model. In principle, this is just an instance of camera pose estimation, but the scale introduces some challenging problems. For one, it makes the correspondence problem very difficult and it is likely that there will be a significant rate of outliers to handle. In this paper we use recent theoretical as well as technical advances to tackle these problems. Many modern cameras and phones have gravitational sensors that allow us to reduce the search space. Further, there are new techniques to efficiently and reliably deal with extreme rates of outliers. We extend these methods to camera pose estimation by using accurate approximations and fast polynomial solvers. Experimental results are given demonstrating that it is possible to reliably estimate the camera pose despite more than 99% of outlier correspondences.
Original languageEnglish
Title of host publicationComputer Vision and Pattern Recognition (CVPR), 2014 IEEE Conference on
PublisherIEEE - Institute of Electrical and Electronics Engineers Inc.
Pages532-539
Number of pages8
DOIs
Publication statusPublished - 2014
EventIEEE Conference on Computer Vision and Pattern Recognition (CVPR 2014), 2014 - Columbus, Ohio, United States
Duration: 2014 Jun 242014 Jun 27

Publication series

Name
ISSN (Print)1063-6919

Conference

ConferenceIEEE Conference on Computer Vision and Pattern Recognition (CVPR 2014), 2014
Country/TerritoryUnited States
CityColumbus, Ohio
Period2014/06/242014/06/27

Subject classification (UKÄ)

  • Computer Vision and Robotics (Autonomous Systems)
  • Mathematics

Free keywords

  • Localization Optimization Polynomial solvers Pose Estimation

Fingerprint

Dive into the research topics of 'Accurate Localization and Pose Estimation for Large 3D Models'. Together they form a unique fingerprint.

Cite this