Results 1  10
of
17
Iterative point matching for registration of freeform curves and surfaces
, 1994
"... A heuristic method has been developed for registering two sets of 3D curves obtained by using an edgebased stereo system, or two dense 3D maps obtained by using a correlationbased stereo system. Geometric matching in general is a difficult unsolved problem in computer vision. Fortunately, in ma ..."
Abstract

Cited by 483 (6 self)
 Add to MetaCart
A heuristic method has been developed for registering two sets of 3D curves obtained by using an edgebased stereo system, or two dense 3D maps obtained by using a correlationbased stereo system. Geometric matching in general is a difficult unsolved problem in computer vision. Fortunately, in many practical applications, some a priori knowledge exists which considerably simplifies the problem. In visual navigation, for example, the motion between successive positions is usually approximately known. From this initial estimate, our algorithm computes observer motion with very good precision, which is required for environment modeling (e.g., building a Digital Elevation Map). Objects are represented by a set of 3D points, which are considered as the samples of a surface. No constraint is imposed on the form of the objects. The proposed algorithm is based on iteratively matching points in one set to the closest points in the other. A statistical method based on the distance distribution is used to deal with outliers, occlusion, appearance and disappearance, which allows us to do subsetsubset matching. A leastsquares technique is used to estimate 3D motion from the point correspondences, which reduces the average distance between points in the two sets. Both synthetic and real data have been used to test the algorithm, and the results show that it is efficient and robust, and yields an accurate motion estimate.
Comparison and evaluation of retrospective intermodality brain image registration techniques
 JOURNAL OF COMPUTER ASSISTED TOMOGRAPHY
, 1997
"... ..."
Retinotopic organization in human visual cortex and the spatial precision of functional MRI
, 1997
"... A method of using functional magnetic resonance imaging (fMRI) to measure retinotopic organization within human cortex is described. The method is based on a visual stimulus that creates a traveling wave of neural activity within retinotopically organized visual areas. We measured the fMRI signal ca ..."
Abstract

Cited by 86 (8 self)
 Add to MetaCart
A method of using functional magnetic resonance imaging (fMRI) to measure retinotopic organization within human cortex is described. The method is based on a visual stimulus that creates a traveling wave of neural activity within retinotopically organized visual areas. We measured the fMRI signal caused by this stimulus in visual cortex and represented the results on images of the #attened cortical sheet. We used the method to measure visual areas and to evaluate the spatial precision of fMRI. Specifically, we: 1) identified the borders between several retinotopically organized visual areas in the posterior occipital lobe, 2) measured the function relating cortical position to visual field eccentricity within area V1, 3) localized activity to within 1.1 mm of visual cortex, and 4) estimated the spatial resolution of the fMRI signal and found that signal falls to 60 percent at a spatial frequency of 1 cycle per 9 mm of visual cortex. This spatial resolution is consistent with a linespread w...
A Framework for Uncertainty and Validation of 3D Registration Methods based on Points and Frames
 Int. Journal of Computer Vision
, 1997
"... In this paper, we propose and analyze several methods to estimate a rigid transformation from a set of 3D matched points or matched frames, which are important features in geometric algorithms. We also develop tools to predict and verify the accuracy of these estimations. The theoretical contributi ..."
Abstract

Cited by 75 (23 self)
 Add to MetaCart
In this paper, we propose and analyze several methods to estimate a rigid transformation from a set of 3D matched points or matched frames, which are important features in geometric algorithms. We also develop tools to predict and verify the accuracy of these estimations. The theoretical contributions are: an intrinsic model of noise for transformations based on composition rather than addition; a unified formalism for the estimation of both the rigid transformation and its covariance matrix for points or frames correspondences, and a statistical validation method to verify the error estimation, which applies even when no "ground truth" is available. We analyze and demonstrate on synthetic data that our scheme is well behaved. The practical contribution of the paper is the validation of our transformation estimation method in the case of 3D medical images, which shows that an accuracy of the registration far below the size of a voxel can be achieved, and in the case of protein substructure matching, where frame features drastically improve both selectivity and complexity. 1.
Validation of 3D Registration Methods based on Points and Frames
 IN PROCEEDINGS OF THE 5TH INT. CONF ON COMP. VISION (ICCV'95
, 1995
"... In this paper, we propose a new method to estimate a rigid transform from a set of 3D matched points or matched frames, and we concentrate on the analysis of the uncertainty of the estimated transform. The theoretical contributions are an intrinsic model of noise for transformations based on comp ..."
Abstract

Cited by 17 (8 self)
 Add to MetaCart
In this paper, we propose a new method to estimate a rigid transform from a set of 3D matched points or matched frames, and we concentrate on the analysis of the uncertainty of the estimated transform. The theoretical contributions are an intrinsic model of noise for transformations based on composition rather than addition, a unified formalism for the estimation of both the rigid transform and its covariance matrix for points or frames correspondences, and also a statistical validation method to verify the error estimation, which applies even when no "ground truth" is available. The practical contribution is the validation of our transform estimation method in the case of 3D medical images, which shows that a precision of the registration, far below the size of a voxel, can be achieved.
Performance Assessment by Resampling: Rigid Motion Estimators Bogdan Matei
 IN K.W. Bowyer and P.J. Phillips, Empirical Evaluation Techniques in Computer Vision, IEEE
, 1998
"... Quantitative assessment of performance in image understanding tasks with real data is difficult since the data is complex and the different computational modules most often interact. Employing modern statistical techniques we have developed a set of numerical tools which provide rigorous performance ..."
Abstract

Cited by 10 (3 self)
 Add to MetaCart
Quantitative assessment of performance in image understanding tasks with real data is difficult since the data is complex and the different computational modules most often interact. Employing modern statistical techniques we have developed a set of numerical tools which provide rigorous performance measures derived solely from the given input. Covariance matrices and confidence intervals are computed for the estimated parameters and individually for the corrected data points. As an example, the proposed methodology is applied to compare rigid motion estimators. 1: Performance assessment in image understanding The lack of universally accepted, rigorous performance assessment methodology is considered by many as one of the major bottlenecks of progress in image understanding. In a recent paper Christensen and Forstner [7] discuss several objections against the widespread use of evaluation techniques. Most of these objections are well justified.
Absolute Orientation from Uncertain Point Data: A Unified Approach
 Proc. IEEE Conf. on Computer Vision and Pattern Recognition
, 1991
"... published in CVPR 92 A general and exible method for fusing and integrating di erent 2D and 3D measurements for pose estimation is proposed. The 2D measured data is viewed as3D data with in nite uncertainty in a particular direction. The method is implemented using Kalman ltering, it is robust and e ..."
Abstract

Cited by 8 (0 self)
 Add to MetaCart
published in CVPR 92 A general and exible method for fusing and integrating di erent 2D and 3D measurements for pose estimation is proposed. The 2D measured data is viewed as3D data with in nite uncertainty in a particular direction. The method is implemented using Kalman ltering, it is robust and easily parallizable. (Keywords: sensor fusion, Kalman lter, parallel algorithms, pose estimation, model based)
A Stereovision System for a Planetary Rover: Calibration, Correlation, Registration, and Fusion
, 1996
"... This paper describes a complete stereovision system, which was originally developed for planetary applications, but can be used for other applications such as object modeling. A new effective onsite calibration technique has been developed, which can make use of the information from the surrounding ..."
Abstract

Cited by 6 (0 self)
 Add to MetaCart
This paper describes a complete stereovision system, which was originally developed for planetary applications, but can be used for other applications such as object modeling. A new effective onsite calibration technique has been developed, which can make use of the information from the surrounding environment as well as the information from the calibration apparatus. A correlationbased stereo algorithm is used, which can produce sufficient dense range maps with an algorithmic structure for fast implementations. A technique based on iterative closestpoint matching has been developed for registration of successive depth maps and computation of the displacements between successive positions. A statistical method based on the distance distribution is integrated into this registration technique, which allows us to deal with the important problems such as outliers, occlusion, appearance and disappearance. Finally, the registered maps are expressed in the same coordinate system and are fu...
Reconstruction of Sculptured Surface Using Coordinate Measuring Machines
 ADVANCES IN DESIGN AUTOMATION  PROCEEDINGS OF THE 19TH ANNUAL ASME DESIGN AUTOMATION CONFERENCE
, 1993
"... This paper presents a strategy for reverse engineering that uses a coordinate measuring machine to reconstruct threedimensional sculptured surfaces. A rough initial model of the surface is generated manually. An iterative method is then used to refine the surface model until the error is within a d ..."
Abstract

Cited by 2 (0 self)
 Add to MetaCart
This paper presents a strategy for reverse engineering that uses a coordinate measuring machine to reconstruct threedimensional sculptured surfaces. A rough initial model of the surface is generated manually. An iterative method is then used to refine the surface model until the error is within a desired bound. The reverse engineering process is broken down into three phases: data acquisition, surface reconstruction and surface evaluation. For data acquisition, an exhaustivesearch algorithm is used to find a safe probe orientation in the vicinity of the target surface, and a coarse cell decomposition method is followed to manipulate the coordinate measuring machine in its work space. Surfaces are modeled using a Bspline approximation technique. The position difference between the surface model and the measured data is used as a simple criterion to evaluate the quality of the reconstructed surface model. Several