Camera Pose Estimation Based on PnL With a Known Vertical Direction

Louis Lecrosnier, Rémi Boutteau, Pascal Vasseur, Xavier Savatier, Friedrich Fraundorfer

Publikation: Beitrag in einer FachzeitschriftArtikelForschungBegutachtung

Abstract

In this letter, we address the problem of camera pose estimation using two-dimensional (2D) and 3-D line features, also known as PnL (Perspective-n-Line) with a known vertical direction. The minimal number of line correspondences required to estimate the complete camera pose is 3 (P3L) in the general case, yielding to a minimum of eight possible solutions. Prior knowledge of the vertical direction, such as provided by common sensors (e.g., inertial measurement unit, or IMU), reduces the problem to a 4 DoF problem and outputs a single solution. We benefit this fact to decouple the remaining rotation estimation and the translation estimation and we present a twofold contribution: First, we present a linear formulation of the PnL problem in Plücker lines coordinates with a known vertical direction, including a Gauss–Newton-based orientation and location refinement to compensate IMU sensor noise. Second, we …
Originalspracheenglisch
Seiten (von - bis)3852-3859
FachzeitschriftIEEE Robotics and Automation Letters
Jahrgang4
Ausgabenummer4
DOIs
PublikationsstatusVeröffentlicht - 2019

Fingerprint

Cameras
Units of measurement
Sensors

Dies zitieren

Camera Pose Estimation Based on PnL With a Known Vertical Direction. / Lecrosnier, Louis; Boutteau, Rémi ; Vasseur, Pascal; Savatier, Xavier; Fraundorfer, Friedrich.

in: IEEE Robotics and Automation Letters, Jahrgang 4, Nr. 4, 2019, S. 3852-3859.

Publikation: Beitrag in einer FachzeitschriftArtikelForschungBegutachtung

Lecrosnier, Louis ; Boutteau, Rémi ; Vasseur, Pascal ; Savatier, Xavier ; Fraundorfer, Friedrich. / Camera Pose Estimation Based on PnL With a Known Vertical Direction. in: IEEE Robotics and Automation Letters. 2019 ; Jahrgang 4, Nr. 4. S. 3852-3859.
@article{a92d8a6fae60405490cf1a38d28ed14b,
title = "Camera Pose Estimation Based on PnL With a Known Vertical Direction",
abstract = "In this letter, we address the problem of camera pose estimation using two-dimensional (2D) and 3-D line features, also known as PnL (Perspective-n-Line) with a known vertical direction. The minimal number of line correspondences required to estimate the complete camera pose is 3 (P3L) in the general case, yielding to a minimum of eight possible solutions. Prior knowledge of the vertical direction, such as provided by common sensors (e.g., inertial measurement unit, or IMU), reduces the problem to a 4 DoF problem and outputs a single solution. We benefit this fact to decouple the remaining rotation estimation and the translation estimation and we present a twofold contribution: First, we present a linear formulation of the PnL problem in Pl{\"u}cker lines coordinates with a known vertical direction, including a Gauss–Newton-based orientation and location refinement to compensate IMU sensor noise. Second, we …",
author = "Louis Lecrosnier and R{\'e}mi Boutteau and Pascal Vasseur and Xavier Savatier and Friedrich Fraundorfer",
year = "2019",
doi = "10.1109/LRA.2019.2929982",
language = "English",
volume = "4",
pages = "3852--3859",
journal = "IEEE Robotics and Automation Letters",
issn = "2377-3766",
publisher = "Institute of Electrical and Electronics Engineers",
number = "4",

}

TY - JOUR

T1 - Camera Pose Estimation Based on PnL With a Known Vertical Direction

AU - Lecrosnier, Louis

AU - Boutteau, Rémi

AU - Vasseur, Pascal

AU - Savatier, Xavier

AU - Fraundorfer, Friedrich

PY - 2019

Y1 - 2019

N2 - In this letter, we address the problem of camera pose estimation using two-dimensional (2D) and 3-D line features, also known as PnL (Perspective-n-Line) with a known vertical direction. The minimal number of line correspondences required to estimate the complete camera pose is 3 (P3L) in the general case, yielding to a minimum of eight possible solutions. Prior knowledge of the vertical direction, such as provided by common sensors (e.g., inertial measurement unit, or IMU), reduces the problem to a 4 DoF problem and outputs a single solution. We benefit this fact to decouple the remaining rotation estimation and the translation estimation and we present a twofold contribution: First, we present a linear formulation of the PnL problem in Plücker lines coordinates with a known vertical direction, including a Gauss–Newton-based orientation and location refinement to compensate IMU sensor noise. Second, we …

AB - In this letter, we address the problem of camera pose estimation using two-dimensional (2D) and 3-D line features, also known as PnL (Perspective-n-Line) with a known vertical direction. The minimal number of line correspondences required to estimate the complete camera pose is 3 (P3L) in the general case, yielding to a minimum of eight possible solutions. Prior knowledge of the vertical direction, such as provided by common sensors (e.g., inertial measurement unit, or IMU), reduces the problem to a 4 DoF problem and outputs a single solution. We benefit this fact to decouple the remaining rotation estimation and the translation estimation and we present a twofold contribution: First, we present a linear formulation of the PnL problem in Plücker lines coordinates with a known vertical direction, including a Gauss–Newton-based orientation and location refinement to compensate IMU sensor noise. Second, we …

U2 - 10.1109/LRA.2019.2929982

DO - 10.1109/LRA.2019.2929982

M3 - Article

VL - 4

SP - 3852

EP - 3859

JO - IEEE Robotics and Automation Letters

JF - IEEE Robotics and Automation Letters

SN - 2377-3766

IS - 4

ER -