Scale-Adaptive ICP

This paper introduces an improved approach for the volume data registration of human retina. Volume data registration refers to calculating out a near-optimal transformation between two volumes with overlapping region and stitching them together. Iterative closest point (ICP) algorithm is a registration method that deals with registration between points. Classical ICP is time consuming and often traps in local minimum when the overlapping region is not big enough. Optical Coherence Tomography (OCT) volume data are several separate, partially overlapping tiles. To stitch them together is a technology in computer aided diagnosis. In this paper, a new 3D registration algorithm based on improved ICP is presented. First, the Canny edge detector is applied to generate the point cloud set of OCT images. After the detection step, an initial registration method based on the feature points of the point cloud is proposed to determine an initial transformation matrix by using singular value decomposition (SVD) method. Then, an improved ICP method is presented to accomplish fine registration. Corresponding point in the point cloud is weighted to reduce the iteration times of ICP algorithm. Finally, M-estimation is used as the objective function to decrease the impact of outliers. This registration algorithm is used to process human retinal OCT volume pairs that contain an overlapping region of 75ź?ź500ź?ź375 voxels approximately. Then a comparative experiment is conducted on some public-available datasets. The experimental results show that the proposed method outperforms the classical method.

This paper deals with the problem of 3D alignment of faces in the presence of some facial expression changes. The data are three dimensional and obtained using a laser scanner. Our approach is based on the differential geometry of the surface and computes the intrinsic local fiducial points on the surface and on curves that reside on the surface. Because these fiducial points are local, they allow partial alignment, where part of the face is viewed. Moreover, since the fiducial points are relatively affine‐invariant to local affine transformations, they allow matching and alignment when facial expression changes affect part of the face. A fast, noniterative alignment procedure is presented in this paper that establishes reliable correspondences between fiducial points without any prior knowledge of the overall nonlinear global transformation that takes place after the changes in facial expressions. This is achieved through the construction of a set of ordered novel absolute local affine invariants. With enough fiducial points set as correspondents, the overall nonlinear transformation is computed and the face before and after the transformation are aligned. For comparison, we also compare the alignment performance of our method with that of the iterative closest point (ICP) method and the coherent point drift (CPD) method which is based on the Gaussian mixture model by looking at the between‐to‐within variation (signal‐to‐noise ratio, SNR) between these two classes (match vs nonmatch) based on the average alignment errors for the genuine matches and nonmatches, normalized by their respective within variation, and by running it on the 3D face database GavabDB. The separation between the true match and nonmatch is best for our zero‐torsion method (SNR of 1.2), followed by the ICP method (SNR of 0.37)and the CPD method (SNR of 0.15). It is interesting to observe here that, although the alignment error for the true match cases is lowest for the CPD method, it is also extremely low for the nonmatch cases, which would mean that the method would force a query into a wrong face, yielding poor specificity (lots of false alarms). © 2013 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

Registration of two three-dimensional (3-D) point sets is a fundamental problem of 3-D shape measurement and modeling pipeline. This paper investigates the automatic pair-wise method to register partially overlapped range images generated by self-developed fringe pattern profilometry (FPP) system. The method is based on the classic iterative closest point (ICP) algorithm but combined with several extensions to adapt to the experimental data. Firstly, the distance function for correspondence finding is modified to be the weighted linear combination of positions and Euclidean invariant features for improving the probability of convergence. In addition, outliers can be discarded through robust statistics and adaptive thresholding of weighted distances between corresponding point pairs. Both artificial and real data are used to test the proposed method. In the ideal noise-free conditions, the experimental results illustrate that it converges to the global minima. The experimental results also show that the proposed method increases the possibility of global convergence when deal with partially overlapped range images.

The 3D reconstruction and modeling of face, objects and surrounding environment etc., has gained significant attention from the research community in recent days. Availability of depth data from various active and passive sensors have accelerated the work in 3D domain. It finds various applications in robotics, Human Computer Interaction (HCI), medical analysis etc. The major challenge in this area is accurate and real time 3D pose estimation which forms the basis for alignment of multi-view point cloud data. Noise in the depth data due to sensor irregularities significantly affects the accuracy of pose estimation. The main idea of this paper is to compute 2D A-KAZE feature correspondence and map to 3D for obtaining more reliable 3D sparse points. The 2D-3D correspondence is used to estimate accurate pose by formulating a graph optimization problem. The experimental results of this approach is evaluated against well known Iterative Closest Point (ICP) and Singular Value Decomposition (SVD) methods for point cloud registration on face and object data obtained using stereo camera. The results shows that the algorithm is able to estimate the accurate pose with least alignment error as compared with ICP and SVD methods.

In this paper, we present a new variant of ICP (iterative closest point) algorithm based on optical flow for finding better 3D point correspondence and computing six-degrees of freedom (6-DOF) pose of the rigid objects. Our approach integrates optical flow algorithm to ICP method, and proposed method gets advantage of higher resolution of the RGB sensor and depth information of the depth sensor. Fusing color and depth sensors improves pose estimation results. We set ground truth with Vicon motion tracker system. And, our approach is tested on a rigid object and compared to ICP method. Results have shown that optical flow and ICP method computes better rigid object pose estimation and lower error against ICP based pose estimation with only depth positions.

This research presents an algorithm, Rigid Super Curves (RSC), to solve the problem of registering two sets of digitized skulls data under a rigid transformation using crest lines. The method that restitutes the rigid transformation between two sets of fully matched curves is propounded. RSC exploits the non-ambiguity of B-Spline representation of super-curves whilst overcoming the inability of super-curves to restore rigid transformations. A further contribution of this study is a two-stage algorithm based on RSC which registers two sets of partially matched curves under a rigid transformation. The algorithm improves the robustness over feature based methods by considering the structure rather than individual points of the curve. Experimental results on CT scanned skull data show that proposed algorithm is more robust and accurate at restoring the rigid transformation between two sets of crest line data compared with Iterated Closest Point and Super Curves methods.

The registration of point clouds in a three-dimensional space is an important task in many areas of computer vision, including robotics and autonomous driving. The purpose of registration is to find a rigid geometric transformation to align two point clouds. The registration problem can be affected by noise and partiality (two point clouds only have a partial overlap). The Iterative Closed Point (ICP) algorithm is a common method for solving the registration problem. Recently, artificial neural networks have begun to be used in the registration of point clouds. The drawback of ICP and other registration algorithms is the possible convergence to a local minimum. Thus, an important characteristic of a registration algorithm is the ability to avoid local minima. In this paper, we propose an ICP-type registration algorithm (λ-ICP) that uses a multiparameter functional (λ-functional). The proposed λ-ICP algorithm generalizes the NICP algorithm (normal ICP). The application of the λ-functional requires a consistent choice of the eigenvectors of the covariance matrix of two point clouds. The paper also proposes an algorithm for choosing the directions of eigenvectors. The performance of the proposed λ-ICP algorithm is compared with that of a standard point-to-point ICP and neural network Deep Closest Points (DCP).

3D quality inspection is widely applied in many industrial fields including mould design, automotive and blade manufacturing, etc. A commonly used method is to obtain the point cloud of the inspected object and make a comparison between the point cloud and the corresponding CAD model or template. Thus, it is important to align the point cloud with the template first and foremost. Moreover, for the purpose of automatization of quality inspection, this alignment process is expected to be completed without manual interference. In this paper, we propose to combine the particle swarm optimization (PSO) with iterative closest point (ICP) algorithm to achieve the automated point cloud alignment. The combination of the two algorithms can achieve a balance between the alignment speed and accuracy, and avoid the local optimal caused by bad initial position of the point cloud.

Localization and mapping are key requirements for autonomous mobile systems to perform navigation and interaction tasks. Iterative Closest Point (ICP) is widely applied for LiDAR scan-matching in the robotic community. In addition, the standard ICP algorithm only considers geometric information when iteratively searching for the nearest point. However, ICP individually cannot achieve accurate point-cloud registration performance in challenging environments such as dynamic environments and highways. Moreover, the computation of searching for the closest points is an expensive step in the ICP algorithm, which is limited to meet real-time requirements, especially when dealing with large-scale point-cloud data. In this paper, we propose a segment-based scan-matching framework for six degree-of-freedom pose estimation and mapping. The LiDAR generates a large number of ground points when scanning, but many of these points are useless and increase the burden of subsequent processing. To address this problem, we first apply an image-based ground-point extraction method to filter out noise and ground points. The point cloud after removing the ground points is then segmented into disjoint sets. After this step, a standard point-to-point ICP is applied into to calculate the six degree-of-freedom transformation between consecutive scans. Furthermore, once closed loops are detected in the environment, a 6D graph-optimization algorithm for global relaxation (6D simultaneous localization and mapping (SLAM)) is employed. Experiments based on publicly available KITTI datasets show that our method requires less runtime while at the same time achieves higher pose estimation accuracy compared with the standard ICP method and its variants.

We propose a methodology to register medical images of carotid arteries from tracked freehand sweep B-Mode ultrasound (US) and magnetic resonance imaging (MRI) acquisitions. Successful registration of US and MR images will allow a multimodal analysis of atherosclerotic plaque in the carotid artery. The main challenge is the difference in the positions of the patient’s neck during the examinations. While in MRI the patient’s neck remains in a natural position, in US the neck is slightly bent and rotated. Moreover, the image characteristics of US and MRI around the carotid artery are very different. Our technique uses the estimated centerlines of the common, internal and external carotid arteries in each modality as landmarks for registration. For US, we used an algorithm based on a rough lumen segmentation obtained by robust ellipse fitting to estimate the lumen centerline. In MRI, we extract the centerline using a minimum cost path approach in which the cost is defined by medialness and an intensity based similarity term. The two centerlines are aligned by an iterative closest point (ICP) algorithm, using rigid and thin-plate spline transformation models. The resulting point correspondences are used as a soft constraint in a subsequent intensity-based registration, optimizing a weighted sum of mutual information between the US and MRI and the Euclidean distance between corresponding points. Rigid and B-spline transformation models were used in this stage. Experiments were performed on datasets from five healthy volunteers. We compared different registration approaches, in order to evaluate the necessity of each step, and to establish the optimum algorithm configuration. For the validation, we used the Dice similarity index to measure the overlap between lumen segmentations in US and MRI.

Iterative Closest Point (ICP) solves the rigid point cloud registration problem iteratively in two steps: (1) make hard assignments of spatially closest point correspondences, and then (2) find the least-squares rigid transformation. The hard assignments of closest point correspondences based on spatial distances are sensitive to the initial rigid transformation and noisy/outlier points, which often cause ICP to converge to wrong local minima. In this paper, we propose the RPM-Net -- a less sensitive to initialization and more robust deep learning-based approach for rigid point cloud registration. To this end, our network uses the differentiable Sinkhorn layer and annealing to get soft assignments of point correspondences from hybrid features learned from both spatial coordinates and local geometry. To further improve registration performance, we introduce a secondary network to predict optimal annealing parameters. Unlike some existing methods, our RPM-Net handles missing correspondences and point clouds with partial visibility. Experimental results show that our RPM-Net achieves state-of-the-art performance compared to existing non-deep learning and recent deep learning methods. Our source code is available at the project website https://github.com/yewzijian/RPMNet .

We present an algorithm that simultaneously identifies the structure and learns the kinematic model of an articulated body from point cloud data and joint angle data with minimal prior assumptions. The robot arm is represented by a point cloud. The structure is identified by segmenting the point cloud to subsets corresponding to different links according to a minimum matching error criterion that includes spatial continuity. The kinematic model is represented as a set of rigid transformations, which are learned by a stochastic gradient decent-based iterative closest point (ICP) algorithm. The learned transformations are then used to infer the structural relationship between segments. We validate the experiment on both synthetic data and data acquired by a Kinect depth sensor viewing a robot arm moving three of its joints. Our results demonstrate that point cloud can be correctly segmented into different rigid parts; the rigid transformations associated with each part can be learned and the kinematic structure can be correctly inferred. Our proposed approach makes a step toward a unified framework for robot self-discovery and learning forward kinematic models purely with data from depth sensors and joint angle encoders.

In this note, we propose an approach to initialize the Iterative Closest Point (ICP) algorithm to match unlabelled point clouds related by rigid transformations. The method is based on matching the ellipsoids defined by the points' covariance matrices and then testing the various principal half-axes matchings that differ by elements of a finite reflection group. We derive bounds on the robustness of our approach to noise and numerical experiments confirm our theoretical findings.

Keypoint based autonomous registration of terrestrial laser point-clouds

Shape analysis finds many important applications in shape understanding, matching and retrieval. Among the various shape analysis methods, spectral shape analysis aims to study the spectrum of the Laplace–Beltrami operator of some well‐designed shape‐dependent equations and obtain a spectral shape descriptor that can in turn be used for shape analysis purposes. The success of such approaches depends greatly on the discriminating power of a shape descriptor. On the other hand, the medial axis of a shape is widely known for its complete shape representation. It is sensitive to small perturbation of the boundary of a shape which often poses difficulty in its effective use for shape analysis. In this paper, we propose a new spectral shape descriptor, called the medial axis spectrum for 2D shapes, which directly applies spectral analysis to the medial axes of the shapes. We extend the Laplace–Beltrami operator onto the medial axis, and take the solution to an extended Laplacian eigenvalue problem defined on the axis as the medial axis spectrum. The medial axis spectrum is robust in the presence of shape boundary noise, and is invariant under rigid transformations, uniform scaling and isometry of the medial axis. We demonstrate these benefits of such a medial axis spectrum representation through extensive experiments. The medial axis spectrum is further used for 2D shape retrieval, and its superiority over previous work is shown by comparison.