A comprehensive survey on point cloud registration

Huang, Xiaoshui; Mei, Guofeng; Zhang, Jian; Abbas, Rana

doi:10.48550/arxiv.2103.02690

Cited by 70 publications

(87 citation statements)

References 116 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The direct idea of point cloud registration is to find the correspondences first and estimate the transformation matrix based on the correspondences [11]. The iterative closest point (ICP) [2] is the widely acknowledged algorithm.…”

Section: Correspondence-based Registrationmentioning

confidence: 99%

“…Recently, point cloud registration from heterogeneous sensors has attracted more research attention. We use our trained model from ModelNet40 to test on the cross-source point clouds from Kinect and RGB camera (reconstructed using VSFM [21]) [11]. Figure 7 shows one example that the proposed algorithm can achieve accurate registration results on the cross-source point clouds.…”

Section: Cross-source Point Cloud Registrationmentioning

confidence: 99%

“…Point cloud registration is the cornerstone of numerous 3D computer vision tasks, such as 3D matching, 3D search, * Equal contribution † Corresponding author and 3D localization [11]. Given two point clouds, most existing registration methods [2,5,12,24,25] use a pipeline consisting of four separate stages: keypoint detection, feature extraction, correspondence search and transformation estimation.…”

Section: Introductionmentioning

confidence: 99%

“…The use of a keypoint detection reduces matching search space, and the resulting sparse correspondences are sufficient for most tasks, e.g., 3D matching and search. However, a keypoint detector may fail to extract enough interest points that are repeatable between point clouds due to various factors such as repetitive structures, low structures, noise, and outliers [11]. Since noise widely exists in point cloud acquisition, repeatable keypoint detection is extremely challenging.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

GenReg: Deep Generative Method for Fast Point Cloud Registration

Huang¹,

Xu²,

Mei³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Accurate and efficient point cloud registration is a challenge because the noise and a large number of points impact the correspondence search. This challenge is still a remaining research problem since most of the existing methods rely on correspondence search. To solve this challenge, we propose a new data-driven registration algorithm by investigating deep generative neural networks to point cloud registration. Given two point clouds, the motivation is to generate the aligned point clouds directly, which is very useful in many applications like 3D matching and search. We design an end-to-end generative neural network for aligned point clouds generation to achieve this motivation, containing three novel components. Firstly, a point multi-perception layer (MLP) mixer (PointMixer) network is proposed to efficiently maintain both the global and local structure information at multiple levels from the self point clouds. Secondly, a feature interaction module is proposed to fuse information from cross point clouds. Thirdly, a parallel and differential sample consensus method is proposed to calculate the transformation matrix of the input point clouds based on the generated registration results. The proposed generative neural network is trained in a GAN framework by maintaining the data distribution and structure similarity. The experiments on both ModelNet40 and 7Scene datasets demonstrate that the proposed algorithm achieves state-of-the-art accuracy and efficiency. Notably, our method reduces 2× in registration error (CD) and 12× running time compared to the state-of-the-art correspondence-based algorithm.

show abstract

Section: Correspondence-based Registrationmentioning

confidence: 99%

Section: Cross-source Point Cloud Registrationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

GenReg: Deep Generative Method for Fast Point Cloud Registration

Huang¹,

Xu²,

Mei³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Second, scaling up point neural networks to finely sampled shapes (N > 10k points) remains a challenging research topic [37,49,135]. Third, the impact of the choice of a specific feature matching method on the performance of deep learning models remains only partially understood [58]. (a) RobOT is equivalent to a nearest neighbor projection subject to mass distribution constraints that make it robust to translations and small deformation.…”

Section: Introductionmentioning

confidence: 99%

Accurate Point Cloud Registration with Robust Optimal Transport

Shen

Feydy

Liu

et al. 2021

Preprint

View full text Add to dashboard Cite

This work investigates the use of robust optimal transport (OT) for shape matching. Specifically, we show that recent OT solvers improve both optimization-based and deep learning methods for point cloud registration, boosting accuracy at an affordable computational cost. This manuscript starts with a practical overview of modern OT theory. We then provide solutions to the main difficulties in using this framework for shape matching. Finally, we showcase the performance of transport-enhanced registration models on a wide range of challenging tasks: rigid registration for partial shapes; scene flow estimation on the Kitti dataset; and nonparametric registration of lung vascular trees between inspiration and expiration. Our OT-based methods achieve state-of-the-art results on Kitti and for the challenging lung registration task, both in terms of accuracy and scalability. We also release PVT1010, a new public dataset of 1,010 pairs of lung vascular trees with densely sampled points. This dataset provides a challenging use case for point cloud registration algorithms with highly complex shapes and deformations. Our work demonstrates that robust OT enables fast pre-alignment and fine-tuning for a wide range of registration models, thereby providing a new key method for the computer vision toolbox. Our code and dataset are available online at: https://github.com/uncbiag/robot.

show abstract

Point Cloud Registration Based on Global and Local Feature Fusion

Yue

Yuan

et al. 2022

IFIP Advances in Information and Communication Technology

View full text Add to dashboard Cite

Cross-modality point cloud registration is confronted with significant challenges due to inherent differences in modalities between different sensors. We propose a cross-modality point cloud registration framework FF-LOGO: a cross-modality point cloud registration method with feature filtering and local-global optimization. The crossmodality feature correlation filtering module extracts geometric transformation-invariant features from cross-modality point clouds and achieves point selection by feature matching. We also introduce a cross-modality optimization process, including a local adaptive key region aggregation module and a global modality consistency fusion optimization module. Experimental results demonstrate that our two-stage optimization significantly improves the registration accuracy of the feature association and selection module. Our method achieves a substantial increase in recall rate compared to the current state-of-the-art methods on the 3DCSR dataset, improving from 40.59% to 75.74%. Our code will be available at https://github.com/wangmohan17/FFLOGO.

show abstract

A comprehensive survey on point cloud registration

Cited by 70 publications

References 116 publications

GenReg: Deep Generative Method for Fast Point Cloud Registration

GenReg: Deep Generative Method for Fast Point Cloud Registration

Accurate Point Cloud Registration with Robust Optimal Transport

Point Cloud Registration Based on Global and Local Feature Fusion

Contact Info

Product

Resources

About