A comprehensive survey on point cloud registration (2021-03-03T00:00:00.000000Z)

TL;DR

This survey conducts a comprehensive survey, including both same-source and cross-source registration methods, and summarizes the connections between optimization-based and deep learning methods, to provide further research insight.

Abstract

Registration is a transformation estimation problem between two point clouds, which has a unique and critical role in numerous computer vision applications. The developments of optimization-based methods and deep learning methods have improved registration robustness and efficiency. Recently, the combinations of optimization-based and deep learning methods have further improved performance. However, the connections between optimization-based and deep learning methods are still unclear. Moreover, with the recent development of 3D sensors and 3D reconstruction techniques, a new research direction emerges to align cross-source point clouds. This survey conducts a comprehensive survey, including both same-source and cross-source registration methods, and summarize the connections between optimization-based and deep learning methods, to provide further research insight. This survey also builds a new benchmark to evaluate the state-of-the-art registration algorithms in solving cross-source challenges. Besides, this survey summarizes the benchmark data sets and discusses point cloud registration applications across various domains. Finally, this survey proposes potential research directions in this rapidly growing field.

Authors

Xiaoshui Huang

1 papers

Guofeng Mei

1 papers

Jian Zhang

1 papers

TL;DR

Abstract

Authors

References126 items

An application independent review of multimodal 3D registration methods

DeepGMR: Learning Latent Gaussian Mixture Models for Registration

Deep learning based point cloud registration: an overview

Global Optimality for Point Set Registration Using Semidefinite Programming

Feature-Metric Registration: A Fast Semi-Supervised Approach for Robust Point Cloud Registration Without Correspondences

Deep Global Registration

RPM-Net: Robust Point Matching Using Learned Features

High Definition 3D Map Creation Using GNSS/IMU/LiDAR Sensor Integration to Support Autonomous Vehicle Navigation

Deep Graph Matching Consensus

TEASER: Fast and Certifiable Point Cloud Registration

Color Point Cloud Registration Based on Supervoxel Correspondence

Procrustes registration of two-dimensional statistical shape models without correspondences

PRNet: Self-Supervised Learning for Partial-to-Partial Registration

Iterative Distance-Aware Similarity Matrix Convolution with Mutual-Supervised Point Elimination for Efficient Point Cloud Registration

Fully Convolutional Geometric Features

ClusterMap Building and Relocalization in Urban Environments for Unmanned Vehicles

Elastic Net Constraint-Based Tensor Model for High-Order Graph Matching

AlignNet-3D: Fast Point Cloud Registration of Partially Observed Objects

SiamesePointNet: A Siamese Point Network Architecture for Learning 3D Shape Descriptor

DeepVCP: An End-to-End Deep Neural Network for Point Cloud Registration

Deep Closest Point: Learning Representations for Point Cloud Registration

Learning to Fuse Local Geometric Features for 3D Rigid Data Matching

3D Local Features for Direct Pairwise Registration

SDRSAC: Semidefinite-Based Randomized Approach for Robust Point Cloud Registration Without Correspondences

3DRegNet: A Deep Neural Network for 3D Point Registration

Non-Rigid Point Set Registration Networks

A Polynomial-time Solution for Robust Registration with Extreme Outlier Rates

PointNetLK: Robust & Efficient Point Cloud Registration Using PointNet

Fast Registration for Cross-Source Point Clouds by using Weak Regional Affinity and Pixel-Wise Refinement

A Review of Point Set Registration: From Pairwise Registration to Groupwise Registration

Analyzing the Stability of Underground Mines Using 3D Point Cloud Data and Discontinuum Numerical Analysis

Extreme Relative Pose Estimation for RGB-D Scans via Scene Completion

LIDAR Point Cloud Registration for Sensing and Reconstruction of Unstructured Terrain

The Perfect Match: 3D Point Cloud Matching With Smoothed Densities

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

Real-Time Point Cloud Alignment for Vehicle Localization in a High Resolution 3D Map

PPF-FoldNet: Unsupervised Learning of Rotation Invariant 3D Local Descriptors

3DFeat-Net: Weakly Supervised Local 3D Features for Point Cloud Registration

Registration of Laser Scanning Point Clouds: A Review

Automated Point Cloud Registration Using Visual and Planar Features for Construction Environments

PPFNet: Global Context Aware Local Features for Robust 3D Point Matching

Open3D: A Modern Library for 3D Data Processing

DS*: Tighter Lifting-Free Convex Relaxations for Quadratic Matching Problems

Guaranteed Outlier Removal for Point Cloud Registration with Correspondences

Underground versatile laser scanning solution

Efficient and Accurate Registration of Point Clouds with Plane to Plane Correspondences

Learning Compact Geometric Features

3D Point Cloud Registration for Localization Using a Deep Neural Network Auto-Encoder

An Automatic Robust Point Cloud Registration on Construction Sites

Octree Generating Networks: Efficient Convolutional Architectures for High-resolution 3D Outputs

PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation

Alternating Direction Graph Matching

OctNet: Learning Deep 3D Representations at High Resolutions

A Coarse-to-Fine Algorithm for Registration in 3D Street-View Cross-Source Point Clouds

Fast Global Registration

Relative Scale Estimation and 3D Registration of Multi-Modal Geometry Using Growing Least Squares

Factorized Graph Matching

A Systematic Approach for Cross-Source Point Cloud Registration by Preserving Macro and Micro Structures

UAVs for coastal surveying

Point registration via efficient convex relaxation

Go-ICP: A Globally Optimal Solution to 3D ICP Point-Set Registration

Applications of point cloud technology in geomechanical characterization, analysis and predictive modeling

Closed-form solutions for estimating a rigid motion from plane correspondences extracted from point clouds

3DMatch: Learning Local Geometric Descriptors from RGB-D Reconstructions

A Review of Point Cloud Registration Algorithms for Mobile Robotics

Iterative Most-Likely Point Registration (IMLP): A Robust Algorithm for Computing Optimal Shape Alignment

A Generative Model for the Joint Registration of Multiple Point Sets

SHOT: Unique signatures of histograms for surface and texture description

3D ShapeNets: A deep representation for volumetric shapes

Point Cloud Encoding for 3D Building Model Retrieval

Street view cross-sourced point cloud matching and registration

Sparse Iterative Closest Point

A Theory of Minimal 3D Point to 3D Plane Registration and Its Generalization

The graph matching problem

Kintinuous: Spatially Extended KinectFusion

Group-Wise Registration of Point Sets for Statistical Shape Models