Long-term 3D Localization and Pose from Semantic Labellings
Paper in proceeding, 2017

One of the major challenges in camera pose estimation and 3D localization is identifying features that are approximately invariant across seasons and in different weather and lighting conditions. In this paper, we present a method for performing accurate and robust six degrees-of-freedom camera pose estimation based only on the pixelwise semantic labelling of a single query image. Localization is performed using a sparse 3D model consisting of semantically labelled points and curves, and an error function based on how well these project onto corresponding curves in the query image is developed. The method is evaluated on the recently released Oxford Robotcar dataset, showing that by minimizing this error function, the pose can be recovered with decimeter accuracy in many cases.

Author

Carl Toft

Chalmers, Signals and Systems, Signal Processing and Biomedical Engineering

Carl Olsson

Chalmers, Signals and Systems, Signal Processing and Biomedical Engineering

Lund University

Fredrik Kahl

Lund University

Chalmers, Signals and Systems, Signal Processing and Biomedical Engineering

IEEE International Conference on Computer Vision Workshops

2473-9936 (ISSN)

Vol. 2018-January 650-659
978-1-5386-1034-3 (ISBN)

16th IEEE International Conference on Computer Vision (ICCV)
Venice, Italy,

Subject Categories

Computational Mathematics

Computer Vision and Robotics (Autonomous Systems)

Medical Image Processing

DOI

10.1109/ICCVW.2017.83

More information

Latest update

7/12/2024