Direct pose estimation for planar objects

Po Chen Wu, Hung Yu Tseng, Ming Hsuan Yang, Shao Yi Chien

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Estimating six degrees of freedom poses of a planar object from images is an important problem with numerous applications ranging from robotics to augmented reality. While the state-of-the-art Perspective-n-Point algorithms perform well in pose estimation, the success hinges on whether feature points can be extracted and matched correctly on target objects with rich texture. In this work, we propose a two-step robust direct method for six-dimensional pose estimation that performs accurately on both textured and textureless planar target objects. First, the pose of a planar target object with respect to a calibrated camera is approximately estimated by posing it as a template matching problem. Second, each object pose is refined and disambiguated using a dense alignment scheme. Extensive experiments on both synthetic and real datasets demonstrate that the proposed direct pose estimation algorithm performs favorably against state-of-the-art feature-based approaches in terms of robustness and accuracy under varying conditions. Furthermore, we show that the proposed dense alignment scheme can also be used for accurate pose tracking in video sequences.

Original languageEnglish
Pages (from-to)50-66
Number of pages17
JournalComputer Vision and Image Understanding
Volume172
DOIs
Publication statusPublished - 2018 Jul

Fingerprint

Template matching
Augmented reality
Hinges
Robotics
Textures
Cameras
Experiments

All Science Journal Classification (ASJC) codes

  • Software
  • Signal Processing
  • Computer Vision and Pattern Recognition

Cite this

Wu, Po Chen ; Tseng, Hung Yu ; Yang, Ming Hsuan ; Chien, Shao Yi. / Direct pose estimation for planar objects. In: Computer Vision and Image Understanding. 2018 ; Vol. 172. pp. 50-66.
@article{a3ec7bd70e964506a088f4d923ec31db,
title = "Direct pose estimation for planar objects",
abstract = "Estimating six degrees of freedom poses of a planar object from images is an important problem with numerous applications ranging from robotics to augmented reality. While the state-of-the-art Perspective-n-Point algorithms perform well in pose estimation, the success hinges on whether feature points can be extracted and matched correctly on target objects with rich texture. In this work, we propose a two-step robust direct method for six-dimensional pose estimation that performs accurately on both textured and textureless planar target objects. First, the pose of a planar target object with respect to a calibrated camera is approximately estimated by posing it as a template matching problem. Second, each object pose is refined and disambiguated using a dense alignment scheme. Extensive experiments on both synthetic and real datasets demonstrate that the proposed direct pose estimation algorithm performs favorably against state-of-the-art feature-based approaches in terms of robustness and accuracy under varying conditions. Furthermore, we show that the proposed dense alignment scheme can also be used for accurate pose tracking in video sequences.",
author = "Wu, {Po Chen} and Tseng, {Hung Yu} and Yang, {Ming Hsuan} and Chien, {Shao Yi}",
year = "2018",
month = "7",
doi = "10.1016/j.cviu.2018.03.006",
language = "English",
volume = "172",
pages = "50--66",
journal = "Computer Vision and Image Understanding",
issn = "1077-3142",
publisher = "Academic Press Inc.",

}

Direct pose estimation for planar objects. / Wu, Po Chen; Tseng, Hung Yu; Yang, Ming Hsuan; Chien, Shao Yi.

In: Computer Vision and Image Understanding, Vol. 172, 07.2018, p. 50-66.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Direct pose estimation for planar objects

AU - Wu, Po Chen

AU - Tseng, Hung Yu

AU - Yang, Ming Hsuan

AU - Chien, Shao Yi

PY - 2018/7

Y1 - 2018/7

N2 - Estimating six degrees of freedom poses of a planar object from images is an important problem with numerous applications ranging from robotics to augmented reality. While the state-of-the-art Perspective-n-Point algorithms perform well in pose estimation, the success hinges on whether feature points can be extracted and matched correctly on target objects with rich texture. In this work, we propose a two-step robust direct method for six-dimensional pose estimation that performs accurately on both textured and textureless planar target objects. First, the pose of a planar target object with respect to a calibrated camera is approximately estimated by posing it as a template matching problem. Second, each object pose is refined and disambiguated using a dense alignment scheme. Extensive experiments on both synthetic and real datasets demonstrate that the proposed direct pose estimation algorithm performs favorably against state-of-the-art feature-based approaches in terms of robustness and accuracy under varying conditions. Furthermore, we show that the proposed dense alignment scheme can also be used for accurate pose tracking in video sequences.

AB - Estimating six degrees of freedom poses of a planar object from images is an important problem with numerous applications ranging from robotics to augmented reality. While the state-of-the-art Perspective-n-Point algorithms perform well in pose estimation, the success hinges on whether feature points can be extracted and matched correctly on target objects with rich texture. In this work, we propose a two-step robust direct method for six-dimensional pose estimation that performs accurately on both textured and textureless planar target objects. First, the pose of a planar target object with respect to a calibrated camera is approximately estimated by posing it as a template matching problem. Second, each object pose is refined and disambiguated using a dense alignment scheme. Extensive experiments on both synthetic and real datasets demonstrate that the proposed direct pose estimation algorithm performs favorably against state-of-the-art feature-based approaches in terms of robustness and accuracy under varying conditions. Furthermore, we show that the proposed dense alignment scheme can also be used for accurate pose tracking in video sequences.

UR - http://www.scopus.com/inward/record.url?scp=85048449498&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85048449498&partnerID=8YFLogxK

U2 - 10.1016/j.cviu.2018.03.006

DO - 10.1016/j.cviu.2018.03.006

M3 - Article

AN - SCOPUS:85048449498

VL - 172

SP - 50

EP - 66

JO - Computer Vision and Image Understanding

JF - Computer Vision and Image Understanding

SN - 1077-3142

ER -