Vision based vehicle relocalization in 3D line-feature map using Perspective-n-Line with a known vertical direction

  title={Vision based vehicle relocalization in 3D line-feature map using Perspective-n-Line with a known vertical direction},
  author={Louis Lecrosnier and R{\'e}mi Boutteau and P. Vasseur and X. Savatier and F. Fraundorfer},
  journal={2019 IEEE Intelligent Transportation Systems Conference (ITSC)},
Common approaches for vehicle localization pro-pose to match LiDAR data or 2D features from cameras to a prior 3D LiDAR map. Yet, these methods require both heavy computational power often provided by GPU, and a first rough localization estimate via GNSS to be performed online. Moreover, storing and accessing 3D dense LiDAR maps can be challenging in case of city-wide coverage.In this paper, we address the problem of camera global relocalization in a prior 3D line-feature map from a single… Expand
Globally Optimal Camera Orientation Estimation from Line Correspondences by BnB algorithm
The results show that the proposed algorithm can obtain the optimal camera orientation and are more robust than several state-of-the-art PnL methods. Expand
Research on Traffic Object 3D Pose Estimation Integrated with Prior Knowledge
This thesis aims to investigate the problem of pose estimation of traffic objects from the image sequences, which can be applied in the application of unmanned vehicle technologies or traffic scenario construction, and proposes a central-line based method for reconstructing the road structure and estimating the road pose simultaneously. Expand


Absolute pose estimation from line correspondences using direct linear transformation
This work proposes a novel method 'DLT-Combined-Lines' based on the Direct Linear Transformation (DLT) algorithm, which benefits from a new combination of two existing DLT methods for pose estimation, and is the most accurate method on real world data, outperforming the state-of-the-art by a large margin. Expand
Visual localization within LIDAR maps for automated urban driving
  • R. W. Wolcott, R. Eustice
  • Computer Science
  • 2014 IEEE/RSJ International Conference on Intelligent Robots and Systems
  • 2014
This paper proposes to localize a single monocular camera within a 3D prior ground-map, generated by a survey vehicle equipped with 3D LIDAR scanners, to obtain comparable localization accuracy with significantly cheaper, commodity cameras. Expand
Pose Estimation from Line Correspondences: A Complete Analysis and a Series of Solutions
This paper provides a complete 3D configuration analysis for P3L, which includes the well-known P3P problem as well as several existing analyses as special cases, and proposed linear-formulation-based PnL approaches inspired by their PnP counterparts are proposed. Expand
Multiview Absolute Pose Using 3D – 2D Perspective Line Correspondences and Vertical Direction
  • Nora Horanyi, Z. Kato
  • Computer Science
  • 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)
  • 2017
This paper addresses the problem of estimating the absolute pose of a multiview calibrated perspective camera system from 3D - 2D line correspondences by proposing two solutions which can be used as a minimal solver as well as a least squares solver without reformulation. Expand
Globally optimal pose estimation from line correspondences
This paper introduces a novel approach to estimate the camera's pose by directly solving the corresponding least-squares problem algebraically through the eigendecomposition of a so-called multiplication matrix. Expand
Camera Pose Estimation from Lines using Plücker Coordinates
Correspondences between 3D lines and their 2D images captured by a camera are often used to determine position and orientation of the camera in space. In this work, we propose a novel algebraicExpand
Robust camera pose estimation from unknown or known line correspondences.
A new line feature based pose estimation algorithm is proposed, which iteratively optimizes the objective function in the object space and has strong robustness to noise and outliers and can attain very accurate results efficiently. Expand
Direct Visual Localisation and Calibration for Road Vehicles in Changing City Environments
A large-scale evaluation of a visual localisation method in a challenging city environment that makes use of a map built by combining data from LIDAR and cameras mounted on a survey vehicle to build a dense appearance prior of the environment, which produces a localiser that is robust to significant changes in scene appearance. Expand
Linear Pose Estimation from Points or Lines
A general framework is presented which allows for a novel set of linear solutions to the pose estimation problem for both n points and n lines and compares the results to two other recent linear algorithms, as well as to iterative approaches. Expand
Sensor Pose Estimation from Multi-center Cylindrical Panoramas
This paper shows that there exist linear models for sensor pose estimation under minor geometrical constraints, namely symmetric and leveled panoramas. Expand