A comprehensive survey on point cloud registration (original) (raw)

Challenging the Universal Representation of Deep Models for 3D Point Cloud Registration

arXiv (Cornell University), 2022

Learning universal representations across different applications domain is an open research problem. In fact, finding universal architecture within the same application but across different types of datasets is still unsolved problem too, especially in applications involving processing 3D point clouds. In this work we experimentally test several stateof-the-art learning-based methods for 3D point cloud registration against the proposed non-learning baseline registration method. The proposed method either outperforms or achieves comparable results w.r.t. learning based methods. In addition, we propose a dataset on which learning based methods have a hard time to generalize. Our proposed method and dataset, along with the provided experiments, can be used in further research in studying effective solutions for universal representations. Our source code is available at: github.com/DavidBoja/greedy-grid-search.

PointNetLK: Robust Efficient Point Cloud Registration Using PointNet

2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)

PointNet has revolutionized how we think about representing point clouds. For classification and segmentation tasks, the approach and its subsequent extensions are stateof-the-art. To date, the successful application of PointNet to point cloud registration has remained elusive. In this paper we argue that PointNet itself can be thought of as a learnable "imaging" function. As a consequence, classical vision algorithms for image alignment can be applied on the problem-namely the Lucas & Kanade (LK) algorithm. Our central innovations stem from: (i) how to modify the LK algorithm to accommodate the PointNet imaging function, and (ii) unrolling PointNet and the LK algorithm into a single trainable recurrent deep neural network. We describe the architecture, and compare its performance against state-of-the-art in common registration scenarios. The architecture offers some remarkable properties including: generalization across shape categories and computational efficiency-opening up new paths of exploration for the application of deep learning to point cloud registration. Code and videos are available at https: //github.com/hmgoforth/PointNetLK.

Challenging data sets for point cloud registration algorithms

International Journal of Robotics Research, 2012

Many registration solutions have bloomed lately in the literature. The iterative closest point, for example, could be considered as the backbone of many laser-based localization and mapping systems. Although they are widely used, it is a common challenge to compare registration solutions on a fair base. The main limitation is to overcome the lack of accurate ground truth in current data sets, which usually cover environments only over a small range of organization levels. In computer vision, the Stanford 3D Scanning Repository pushed forward point cloud registration algorithms and object modeling fields by providing high-quality scanned objects with precise localization. We aim at providing similar high-caliber working material to the robotic and computer vision communities but with sceneries instead of objects. We propose 8 point cloud sequences acquired in locations covering the environment diversity that modern robots are susceptible to encounter, ranging from inside an apartment to a woodland area. The core of the data sets consists of 3D laser point clouds for which supporting data (Gravity, Magnetic North and GPS) are given at each pose. A special effort has been made to ensure a global positioning of the scanner within millimeter range precision, independently of environmental conditions. This will allow for the development of improved registration algorithms when mapping challenging environments, such as found in real world situations.

Integrating Deep Semantic Segmentation Into 3-D Point Cloud Registration

IEEE Robotics and Automation Letters

Point cloud registration is the task of aligning 3D scans of the same environment captured from different poses. When semantic information is available for the points, it can be used as a prior in the search for correspondences to improve registration. Semantic-assisted Normal Distributions Transform (SE-NDT) is a new registration algorithm that reduces the complexity of the problem by using the semantic information to partition the point cloud into a set of normal distributions, which are then registered separately. In this paper we extend the NDT registration pipeline by using PointNet, a deep neural network for segmentation and classification of point clouds, to learn and predict per-point semantic labels. We also present the Iterative Closest Point (ICP) equivalent of the algorithm, a special case of Multichannel Generalized ICP. We evaluate the performance of SE-NDT against the state of the art in point cloud registration on the publicly available classification data set Semantic3d.net. We also test the trained classifier and algorithms on dynamic scenes, using a sequence from the public dataset KITTI. The experiments demonstrate the improvement of the registration in terms of robustness, precision and speed, across a range of initial registration errors, thanks to the inclusion of semantic information.

A Coarse-to-Fine Algorithm for Matching and Registration in 3D Cross-Source Point Clouds

IEEE Transactions on Circuits and Systems for Video Technology, 2017

We propose an efficient method to deal with the matching and registration problem found in cross-source point clouds captured by different types of sensors. This task is especially challenging due to the presence of density variation, scale difference, a large proportion of noise and outliers, missing data and viewpoint variation. The proposed method has two stages: in the coarse matching stage, we use the ESF descriptor to select potential K regions from the candidate point clouds for the target. In the fine stage, we propose a scale embedded generative GMM registration method to refine the results from the coarse matching stage. Following the fine stage, both the best region and accurate camera pose relationships between the candidates and target are found. We conduct experiments in which we apply the method to two applications: one is 3D object detection and localization in street-view ourdoor (LiDAR/VSFM) cross-source point clouds, and the other is 3D scene matching and registration in indoor (KinectFusion/VSFM) cross-source point clouds. The experiment results show that the proposed method performs well when compared with the existing methods. It also shows that the proposed method is robust under various sensing techniques such as LiDAR, Kinect and RGB camera.

SAP-Net: A Simple and Robust 3D Point Cloud Registration Network Based on Local Shape Features

Sensors, 2021

Point cloud registration is a key step in the reconstruction of 3D data models. The traditional ICP registration algorithm depends on the initial position of the point cloud. Otherwise, it may get trapped into local optima. In addition, the registration method based on the feature learning of PointNet cannot directly or effectively extract local features. To solve these two problems, this paper proposes SAP-Net, inspired by CorsNet and PointNet++, as an optimized CorsNet. To be more specific, SAP-Net firstly uses the set abstraction layer in PointNet++ as the feature extraction layer and then combines the global features with the initial template point cloud. Finally, PointNet is used as the transform prediction layer to obtain the six parameters required for point cloud registration directly, namely the rotation matrix and the translation vector. Experiments on the ModelNet40 dataset and real data show that SAP-Net not only outperforms ICP and CorsNet on both seen and unseen catego...

Globally optimal registration of noisy point clouds

2019

Registration of 3D point clouds is a fundamental task in several applications of robotics and computer vision. While registration methods such as iterative closest point and variants are very popular, they are only locally optimal. There has been some recent work on globally optimal registration, but they perform poorly in the presence of noise in the measurements. In this work we develop a mixed integer programming-based approach for globally optimal registration that explicitly considers uncertainty in its optimization, and hence produces more accurate estimates. Furthermore, from a practical implementation perspective we develop a multi-step optimization that combines fast local methods with our accurate global formulation. Through extensive simulation and real world experiments we demonstrate improved performance over state-of-the-art methods for various level of noise and outliers in the data as well as for partial geometric overlap.

Registration of point cloud data from a geometric optimization perspective

2004

We propose a framework for pairwise registration of shapes represented by point cloud data (PCD). We assume that the points are sampled from a surface and formulate the problem of aligning two PCDs as a minimization of the squared distance between the underlying surfaces. Local quadratic approximants of the squared distance function are used to develop a linear system whose solution gives the best aligning rigid transform for the given pair of point clouds. The rigid transform is applied and the linear system corresponding to the new orientation is build. This process is iterated until it converges. The point-to-point and the point-to-plane Iterated Closest Point (ICP) algorithms can be treated as special cases in this framework. Our algorithm can align PCDs even when they are placed far apart, and is experimentally found to be more stable than point-to-plane ICP. We analyze the convergence behavior of our algorithm and of point-to-point and point-to-plane ICP under our proposed framework, and derive bounds on their rate of convergence. We compare the stability and convergence properties of our algorithm with other registration algorithms on a variety of scanned data.

Fast and Robust Registration of Partially Overlapping Point Clouds

IEEE Robotics and Automation Letters, 2022

Real-time registration of partially overlapping point clouds has emerging applications in cooperative perception for autonomous vehicles and multi-agent SLAM. The relative translation between point clouds in these applications is higher than in traditional SLAM and odometry applications, which challenges the identification of correspondences and a successful registration. In this paper, we propose a novel registration method for partially overlapping point clouds where correspondences are learned using an efficient point-wise feature encoder, and refined using a graph-based attention network. This attention network exploits geometrical relationships between key points to improve the matching in point clouds with low overlap. At inference time, the relative pose transformation is obtained by robustly fitting the correspondences through sample consensus. The evaluation is performed on the KITTI dataset and a novel synthetic dataset including low-overlapping point clouds with displacements of up to 30m. The proposed method achieves onpar performance with state-of-the-art methods on the KITTI dataset, and outperforms existing methods for low overlapping point clouds. Additionally, the proposed method achieves significantly faster inference times, as low as 410ms, between 5 and 35 times faster than competing methods. Our code and dataset are available at https://github.com/eduardohenriquearnold/fastreg.

Using Extended Measurements and Scene Merging for Efficient and Robust Point Cloud Registration

Point cloud registration is a fundamental building block of many robotic applications. In this paper we describe a system to solve the registration problem, that builds on top of our previous work [1], and that represents an extension to the well known Iterative Closest Point (ICP) algorithm. Our approach combines recent achievements on optimization by using an extended point representation [2] that captures the surface characteristics around the points. Thanks to an effective strategy to search for correspondences, our method can operate on-line and cope with measurements gathered with an heterogeneous set of range and depth sensors. By using an efficient map-merging procedure our approach can quickly update the tracked scene and handle dynamic aspects. We also introduce an approximated variant of our method that runs at twice the speed of our full implementation. Experiments performed on a large publicly available bench-marking dataset show that our approach performs better with respect to other state-of-the art methods. In most of the tests considered, our algorithm has been able to obtain a translational and rotational relative error of respectively ∼1 cm and ∼1 degree.