Robust 2D-3D alignment based on geometrical consistency

Kenji Hara, Yuuki Kabashima, Yumi Iwashita, Ryo Kurazume, Tsutomu Hasegawa

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

This paper presents a new registration algorithm of a 2D image and a 3D geometrical model, which is robust for initial registration errors, for reconstructing a realistic 3D model of indoor scene settings. One of the typical techniques of pose estimation of a 3D model in a 2D image is the method based on the correspondences between 2D photometrical edges and 3D geometrical edges projected on the 2D image. However, for indoor settings, features extracted on the 2D image and jump edges of the geometrical model, which can be extracted robustly, are limited. Therefore, it is difficult to find corresponding edges between the 2D image and the 3D model correctly. For this reason, in most cases, the relative position has to be manually set close to correct position beforehand. To overcome this problem, in the proposed method, firstly the relative pose is roughly estimated by utilizing geometrical consistencies of back-projected 2D photometrical edges on a 3D model. Next, the edge-based method is applied for the precise pose estimation after the above estimation procedure is converged. The performance of the proposed method is successfully demonstrated with some experiments using simulated models of indoor scene settings and actual environments measured by range and image sensors.

Original languageEnglish
Title of host publicationProceedings - 6th International Conference on 3-D Digital Imaging and Modeling, 3DIM 2007
Pages273-280
Number of pages8
DOIs
Publication statusPublished - Dec 1 2007
Event6th International Conference on 3-D Digital Imaging and Modeling, 3DIM 2007 - Montreal, QC, Canada
Duration: Aug 21 2007Aug 23 2007

Publication series

Name3DIM 2007 - Proceedings 6th International Conference on 3-D Digital Imaging and Modeling

Other

Other6th International Conference on 3-D Digital Imaging and Modeling, 3DIM 2007
CountryCanada
CityMontreal, QC
Period8/21/078/23/07

Fingerprint

Image sensors
Experiments

All Science Journal Classification (ASJC) codes

  • Computer Graphics and Computer-Aided Design
  • Computer Science Applications
  • Computer Vision and Pattern Recognition

Cite this

Hara, K., Kabashima, Y., Iwashita, Y., Kurazume, R., & Hasegawa, T. (2007). Robust 2D-3D alignment based on geometrical consistency. In Proceedings - 6th International Conference on 3-D Digital Imaging and Modeling, 3DIM 2007 (pp. 273-280). [4296765] (3DIM 2007 - Proceedings 6th International Conference on 3-D Digital Imaging and Modeling). https://doi.org/10.1109/3DIM.2007.44

Robust 2D-3D alignment based on geometrical consistency. / Hara, Kenji; Kabashima, Yuuki; Iwashita, Yumi; Kurazume, Ryo; Hasegawa, Tsutomu.

Proceedings - 6th International Conference on 3-D Digital Imaging and Modeling, 3DIM 2007. 2007. p. 273-280 4296765 (3DIM 2007 - Proceedings 6th International Conference on 3-D Digital Imaging and Modeling).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Hara, K, Kabashima, Y, Iwashita, Y, Kurazume, R & Hasegawa, T 2007, Robust 2D-3D alignment based on geometrical consistency. in Proceedings - 6th International Conference on 3-D Digital Imaging and Modeling, 3DIM 2007., 4296765, 3DIM 2007 - Proceedings 6th International Conference on 3-D Digital Imaging and Modeling, pp. 273-280, 6th International Conference on 3-D Digital Imaging and Modeling, 3DIM 2007, Montreal, QC, Canada, 8/21/07. https://doi.org/10.1109/3DIM.2007.44
Hara K, Kabashima Y, Iwashita Y, Kurazume R, Hasegawa T. Robust 2D-3D alignment based on geometrical consistency. In Proceedings - 6th International Conference on 3-D Digital Imaging and Modeling, 3DIM 2007. 2007. p. 273-280. 4296765. (3DIM 2007 - Proceedings 6th International Conference on 3-D Digital Imaging and Modeling). https://doi.org/10.1109/3DIM.2007.44
Hara, Kenji ; Kabashima, Yuuki ; Iwashita, Yumi ; Kurazume, Ryo ; Hasegawa, Tsutomu. / Robust 2D-3D alignment based on geometrical consistency. Proceedings - 6th International Conference on 3-D Digital Imaging and Modeling, 3DIM 2007. 2007. pp. 273-280 (3DIM 2007 - Proceedings 6th International Conference on 3-D Digital Imaging and Modeling).
@inproceedings{5d6cf737abd6495290e0200038609e32,
title = "Robust 2D-3D alignment based on geometrical consistency",
abstract = "This paper presents a new registration algorithm of a 2D image and a 3D geometrical model, which is robust for initial registration errors, for reconstructing a realistic 3D model of indoor scene settings. One of the typical techniques of pose estimation of a 3D model in a 2D image is the method based on the correspondences between 2D photometrical edges and 3D geometrical edges projected on the 2D image. However, for indoor settings, features extracted on the 2D image and jump edges of the geometrical model, which can be extracted robustly, are limited. Therefore, it is difficult to find corresponding edges between the 2D image and the 3D model correctly. For this reason, in most cases, the relative position has to be manually set close to correct position beforehand. To overcome this problem, in the proposed method, firstly the relative pose is roughly estimated by utilizing geometrical consistencies of back-projected 2D photometrical edges on a 3D model. Next, the edge-based method is applied for the precise pose estimation after the above estimation procedure is converged. The performance of the proposed method is successfully demonstrated with some experiments using simulated models of indoor scene settings and actual environments measured by range and image sensors.",
author = "Kenji Hara and Yuuki Kabashima and Yumi Iwashita and Ryo Kurazume and Tsutomu Hasegawa",
year = "2007",
month = "12",
day = "1",
doi = "10.1109/3DIM.2007.44",
language = "English",
isbn = "0769529399",
series = "3DIM 2007 - Proceedings 6th International Conference on 3-D Digital Imaging and Modeling",
pages = "273--280",
booktitle = "Proceedings - 6th International Conference on 3-D Digital Imaging and Modeling, 3DIM 2007",

}

TY - GEN

T1 - Robust 2D-3D alignment based on geometrical consistency

AU - Hara, Kenji

AU - Kabashima, Yuuki

AU - Iwashita, Yumi

AU - Kurazume, Ryo

AU - Hasegawa, Tsutomu

PY - 2007/12/1

Y1 - 2007/12/1

N2 - This paper presents a new registration algorithm of a 2D image and a 3D geometrical model, which is robust for initial registration errors, for reconstructing a realistic 3D model of indoor scene settings. One of the typical techniques of pose estimation of a 3D model in a 2D image is the method based on the correspondences between 2D photometrical edges and 3D geometrical edges projected on the 2D image. However, for indoor settings, features extracted on the 2D image and jump edges of the geometrical model, which can be extracted robustly, are limited. Therefore, it is difficult to find corresponding edges between the 2D image and the 3D model correctly. For this reason, in most cases, the relative position has to be manually set close to correct position beforehand. To overcome this problem, in the proposed method, firstly the relative pose is roughly estimated by utilizing geometrical consistencies of back-projected 2D photometrical edges on a 3D model. Next, the edge-based method is applied for the precise pose estimation after the above estimation procedure is converged. The performance of the proposed method is successfully demonstrated with some experiments using simulated models of indoor scene settings and actual environments measured by range and image sensors.

AB - This paper presents a new registration algorithm of a 2D image and a 3D geometrical model, which is robust for initial registration errors, for reconstructing a realistic 3D model of indoor scene settings. One of the typical techniques of pose estimation of a 3D model in a 2D image is the method based on the correspondences between 2D photometrical edges and 3D geometrical edges projected on the 2D image. However, for indoor settings, features extracted on the 2D image and jump edges of the geometrical model, which can be extracted robustly, are limited. Therefore, it is difficult to find corresponding edges between the 2D image and the 3D model correctly. For this reason, in most cases, the relative position has to be manually set close to correct position beforehand. To overcome this problem, in the proposed method, firstly the relative pose is roughly estimated by utilizing geometrical consistencies of back-projected 2D photometrical edges on a 3D model. Next, the edge-based method is applied for the precise pose estimation after the above estimation procedure is converged. The performance of the proposed method is successfully demonstrated with some experiments using simulated models of indoor scene settings and actual environments measured by range and image sensors.

UR - http://www.scopus.com/inward/record.url?scp=47349114896&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=47349114896&partnerID=8YFLogxK

U2 - 10.1109/3DIM.2007.44

DO - 10.1109/3DIM.2007.44

M3 - Conference contribution

AN - SCOPUS:47349114896

SN - 0769529399

SN - 9780769529394

T3 - 3DIM 2007 - Proceedings 6th International Conference on 3-D Digital Imaging and Modeling

SP - 273

EP - 280

BT - Proceedings - 6th International Conference on 3-D Digital Imaging and Modeling, 3DIM 2007

ER -