Scale-invariant registration of monocular stereo images to 3D surface models

Darius Burschka, Ming Li, Russell H Taylor, Gregory Hager

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

Abstract

We present an approach for scale recovery from monocular stereo images of an endoscopic camera with simultaneous registration to dense 3D surface models. We assume the camera motion to be unknown or at least uncertain. An example application is the registration of endoscope images to pre-operative CT scans that allows instrument navigation during surgical procedures. The application field is not restricted to the medical field. It can be extended to registration of monocular video images to laser-based surface reconstructions in, e.g., mobile navigation area or to autonomous aircraft navigation from topological surveys. A novel way for depth estimation from arbitrary camera motion is presented. In this paper, we focus on the robust initialization of the system and on the scale recovery for the reconstructed 3D point clouds with accurate registration to the candidate surfaces extracted from the CT data. We provide experimental validation of the algorithm with data obtained from our experiments with a phantom skull.

Original languageEnglish (US)
Title of host publication2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
Pages2581-2586
Number of pages6
Volume3
StatePublished - 2004
Event2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) - Sendai, Japan
Duration: Sep 28 2004Oct 2 2004

Other

Other2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
CountryJapan
CitySendai
Period9/28/0410/2/04

Fingerprint

Navigation
Cameras
Recovery
Endoscopy
Surface reconstruction
Computerized tomography
Aircraft
Lasers
Experiments

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Burschka, D., Li, M., Taylor, R. H., & Hager, G. (2004). Scale-invariant registration of monocular stereo images to 3D surface models. In 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (Vol. 3, pp. 2581-2586)

Scale-invariant registration of monocular stereo images to 3D surface models. / Burschka, Darius; Li, Ming; Taylor, Russell H; Hager, Gregory.

2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Vol. 3 2004. p. 2581-2586.

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

Burschka, D, Li, M, Taylor, RH & Hager, G 2004, Scale-invariant registration of monocular stereo images to 3D surface models. in 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). vol. 3, pp. 2581-2586, 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Sendai, Japan, 9/28/04.
Burschka D, Li M, Taylor RH, Hager G. Scale-invariant registration of monocular stereo images to 3D surface models. In 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Vol. 3. 2004. p. 2581-2586
Burschka, Darius ; Li, Ming ; Taylor, Russell H ; Hager, Gregory. / Scale-invariant registration of monocular stereo images to 3D surface models. 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Vol. 3 2004. pp. 2581-2586
@inproceedings{7a99aac492ce4377b777524200f4fecc,
title = "Scale-invariant registration of monocular stereo images to 3D surface models",
abstract = "We present an approach for scale recovery from monocular stereo images of an endoscopic camera with simultaneous registration to dense 3D surface models. We assume the camera motion to be unknown or at least uncertain. An example application is the registration of endoscope images to pre-operative CT scans that allows instrument navigation during surgical procedures. The application field is not restricted to the medical field. It can be extended to registration of monocular video images to laser-based surface reconstructions in, e.g., mobile navigation area or to autonomous aircraft navigation from topological surveys. A novel way for depth estimation from arbitrary camera motion is presented. In this paper, we focus on the robust initialization of the system and on the scale recovery for the reconstructed 3D point clouds with accurate registration to the candidate surfaces extracted from the CT data. We provide experimental validation of the algorithm with data obtained from our experiments with a phantom skull.",
author = "Darius Burschka and Ming Li and Taylor, {Russell H} and Gregory Hager",
year = "2004",
language = "English (US)",
isbn = "0780384636",
volume = "3",
pages = "2581--2586",
booktitle = "2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)",

}

TY - GEN

T1 - Scale-invariant registration of monocular stereo images to 3D surface models

AU - Burschka, Darius

AU - Li, Ming

AU - Taylor, Russell H

AU - Hager, Gregory

PY - 2004

Y1 - 2004

N2 - We present an approach for scale recovery from monocular stereo images of an endoscopic camera with simultaneous registration to dense 3D surface models. We assume the camera motion to be unknown or at least uncertain. An example application is the registration of endoscope images to pre-operative CT scans that allows instrument navigation during surgical procedures. The application field is not restricted to the medical field. It can be extended to registration of monocular video images to laser-based surface reconstructions in, e.g., mobile navigation area or to autonomous aircraft navigation from topological surveys. A novel way for depth estimation from arbitrary camera motion is presented. In this paper, we focus on the robust initialization of the system and on the scale recovery for the reconstructed 3D point clouds with accurate registration to the candidate surfaces extracted from the CT data. We provide experimental validation of the algorithm with data obtained from our experiments with a phantom skull.

AB - We present an approach for scale recovery from monocular stereo images of an endoscopic camera with simultaneous registration to dense 3D surface models. We assume the camera motion to be unknown or at least uncertain. An example application is the registration of endoscope images to pre-operative CT scans that allows instrument navigation during surgical procedures. The application field is not restricted to the medical field. It can be extended to registration of monocular video images to laser-based surface reconstructions in, e.g., mobile navigation area or to autonomous aircraft navigation from topological surveys. A novel way for depth estimation from arbitrary camera motion is presented. In this paper, we focus on the robust initialization of the system and on the scale recovery for the reconstructed 3D point clouds with accurate registration to the candidate surfaces extracted from the CT data. We provide experimental validation of the algorithm with data obtained from our experiments with a phantom skull.

UR - http://www.scopus.com/inward/record.url?scp=14044279286&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=14044279286&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:14044279286

SN - 0780384636

VL - 3

SP - 2581

EP - 2586

BT - 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

ER -