Object pose from 2-D to 3-D point and line correspondences

Phong, Thai Quynh; Horaud, Radu; Yassine, Adnan; Tao, Pham Dinh

doi:10.1007/bf01451742

Cited by 131 publications

(49 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…(a) weak perspective pose from line correspondences; that is, the result of the first iteration of the iterative weak perspective algorithm described in this paper; (b) paraperspective pose from line correspondences; that is, the result of the first iteration of the iterative paraperspective algorithm described in this paper; (c) perspective pose from line correspondences obtained with the iterative weak perspective algorithm; (d) perspective pose from point correspondences obtained with the iterative weak perspective algorithm; (e) perspective pose from line correspondences obtained with a nonlinear method [12], and (f) perspective pose from point correspondences obtained with the same nonlinear method.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Iterative Pose Computation from Line Correspondences

Christy

Horaud

1999

Computer Vision and Image Understanding

Self Cite

View full text Add to dashboard Cite

This paper presents a method for estimating the position and orientation of a camera with respect to a known 3-D object from line correspondences. The main idea of the method is to estimate a pose with either a weak perspective or a paraperspective camera model and to improve this pose iteratively. At convergence the result is compatible with a perspective camera model. This iterative improvement of a linear (affine) camera model has already been used for points but has never been extended to lines. Known methods which compute pose from line correspondences deal with a set of nonlinear equations which are solved either in closed-form or using minimization techniques. These methods have to deal with multiple solutions. In contrast our method starts with a solution which is very close to the true solution and converges in very few iterations (typically three to five iterations). The rank analysis of the linear system to be solved at each iteration allows us to characterize geometric configurations which defeat the algorithm.

show abstract

Section: Methodsmentioning

confidence: 99%

“…For more than three lines closed form solutions become unpractical. Both [9] and [12] showed that two classes of methods are possible with lines: rotation then translation and rotation and translation. Both * This work has been supported by Société Aérospatiale and by DGA/DRET.…”

Section: Introductionmentioning

confidence: 99%

Iterative Pose Computation from Line Correspondences

Christy

Horaud

1999

Computer Vision and Image Understanding

Self Cite

View full text Add to dashboard Cite

show abstract

“…Phong et al [14] also present an algorithm for calculating the pose from points and lines, however, their solution is iterative. Liu et al [11] study a two-step process for pose estimation from lines that first iteratively solves for rotation and then linearly for translation.…”

Section: Related Workmentioning

confidence: 99%

Generalised Linear Pose Estimation

Ess¹,

Neubeck²,

Gool

2007

Procedings of the British Machine Vision Conference 2007

View full text Add to dashboard Cite

This paper investigates several aspects of 3D-2D camera pose estimation, aimed at robot navigation in poorly-textured scenes. The major contribution is a fast, linear algorithm for the general case with six or more points. We show how to specialise this to work with only four or five points, which is of utmost importance in a test and hypothesis framework. Our formulation allows for an easy inclusion of lines, as well as the handling of other camera geometries, such as stereo rigs. We also treat the special case of planar motion, a valid restriction for most indoor environments. We conclude the paper with extensive simulated tests and a real test case, which substantiate the algorithm's usability for our application domain.

show abstract

“…Pose estimation can be achieved through corresponding pairs of 2D and 3D features, such as points [8] or points and lines [11]. The accuracy of the pose estimation algorithm is determined by the number of features, the accuracy with which the 2D and 3D coordinates of the features are known and the accuracy of their correspondence.…”

Section: Previous Workmentioning

confidence: 99%

Registration of Video Images to Tomographic Images by Optimising Mutual Information Using Texture Mapping

Clarkson

Rueckert

King

et al. 1999

Medical Image Computing and Computer-Assisted Intervention – MICCAI’99

View full text Add to dashboard Cite

Abstract. In this paper we propose a novel tracking method to update the pose of stereo video cameras with respect to a surface model derived from a 3D tomographic image. This has a number of applications in image guided interventions and therapy. Registration of 2D video images to the pre-operative 3D image provides a mapping between image and physical space and enables a perspective projection of the pre-operative data to be overlaid onto the video image. Assuming an initial registration can be achieved, we propose a method for updating the registration, which is based on image intensity and texture mapping. We performed five experiments on simulated, phantom and volunteer data and validated the algorithm against an accurate gold standard in all three cases. We measured the mean 3D error of our tracking algorithm to be 1.05 mm for the simulation and 1.89 mm for the volunteer data. Visually this corresponds to a good registration.

show abstract

Object pose from 2-D to 3-D point and line correspondences

Cited by 131 publications

References 23 publications

Iterative Pose Computation from Line Correspondences

Iterative Pose Computation from Line Correspondences

Generalised Linear Pose Estimation

Registration of Video Images to Tomographic Images by Optimising Mutual Information Using Texture Mapping

Contact Info

Product

Resources

About