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

PCRNet: Point Cloud Registration Network using PointNet Encoding

2019

PointNet has recently emerged as a popular representation for unstructured point cloud data, allowing application of deep learning to tasks such as object detection, segmentation and shape completion. However, recent works in literature have shown the sensitivity of the PointNet representation to pose misalignment. This paper presents a novel framework that uses the PointNet representation to align point clouds and perform registration for applications such as tracking, 3D reconstruction and pose estimation. We develop a framework that compares PointNet features of template and source point clouds to find the transformation that aligns them accurately. Depending on the prior information about the shape of the object formed by the point clouds, our framework can produce approaches that are shape specific or general to unseen shapes. The shape specific approach uses a Siamese architecture with fully connected (FC) layers and is robust to noise and initial misalignment in data. We perf...

One Framework to Register Them All: PointNet Encoding for Point Cloud Alignment

2019

PointNet has recently emerged as a popular representation for unstructured point cloud data, allowing application of deep learning to tasks such as object detection, segmentation and shape completion. However, recent works in literature have shown the sensitivity of the PointNet representation to pose misalignment. This paper presents a novel framework that uses PointNet encoding to align point clouds and perform registration for applications such as 3D reconstruction, tracking and pose estimation. We develop a framework that compares PointNet features of template and source point clouds to find the transformation that aligns them accurately. In doing so, we avoid computationally expensive correspondence finding steps, that are central to popular registration methods such as ICP and its variants. Depending on the prior information about the shape of the object formed by the point clouds, our framework can produce approaches that are shape specific or general to unseen shapes. Our fr...

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.

PCCT: A Point Cloud Classification Tool to Create 3D Training Data to Adjust and Develop 3D Convnet

The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2019

Point clouds give a very detailed and sometimes very accurate representation of the geometry of captured objects. In surveying, point clouds captured with laser scanners or camera systems are an intermediate result that must be processed further. Often the point cloud has to be divided into regions of similar types (object classes) for the next process steps. These classifications are very time-consuming and cost-intensive compared to acquisition. In order to automate this process step, conventional neural networks (ConvNet), which take over the classification task, are investigated in detail. In addition to the network architecture, the classification performance of a ConvNet depends on the training data with which the task is learned. This paper presents and evaluates the point clould classification tool (PCCT) developed at HCU Hamburg. With the PCCT, large point cloud collections can be semi-automatically classified. Furthermore, the influence of erroneous points in three-dimensional point clouds is investigated. The network architecture PointNet is used for this investigation.

A comprehensive survey on point cloud registration

arXiv: Computer Vision and Pattern Recognition, 2021

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.

CorrNet3D: Unsupervised End-to-end Learning of Dense Correspondence for 3D Point Clouds

2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021

Motivated by the intuition that one can transform two aligned point clouds to each other more easily and meaningfully than a misaligned pair, we propose CorrNet3Dthe first unsupervised and end-to-end deep learning-based framework-to drive the learning of dense correspondence between 3D shapes by means of deformation-like reconstruction to overcome the need for annotated data. Specifically, CorrNet3D consists of a deep feature embedding module and two novel modules called correspondence indicator and symmetric deformer. Feeding a pair of raw point clouds, our model first learns the pointwise features and passes them into the indicator to generate a learnable correspondence matrix used to permute the input pair. The symmetric deformer, with an additional regularized loss, transforms the two permuted point clouds to each other to drive the unsupervised learning of the correspondence. The extensive experiments on both synthetic and real-world datasets of rigid and non-rigid 3D shapes show our CorrNet3D outperforms state-of-the-art methods to a large extent, including those taking meshes as input. CorrNet3D is a flexible framework in that it can be easily adapted to supervised learning if annotated data are available. The source code and pre-trained model will be available at https://github.com/ZENGYIMING-EAMON/CorrNet3D.git.

PCPNetLearning Local Shape Properties from Raw Point Clouds

Computer Graphics Forum, 2018

In this paper, we propose PCPNET, a deep-learning based approach for estimating local 3D shape properties in point clouds. In contrast to the majority of prior techniques that concentrate on global or mid-level attributes, e.g., for shape classification or semantic labeling, we suggest a patch-based learning method, in which a series of local patches at multiple scales around each point is encoded in a structured manner. Our approach is especially well-adapted for estimating local shape properties such as normals (both unoriented and oriented) and curvature from raw point clouds in the presence of strong noise and multi-scale features. Our main contributions include both a novel multi-scale variant of the recently proposed PointNet architecture with emphasis on local shape information, and a series of novel applications in which we demonstrate how learning from training data arising from well-structured triangle meshes, and applying the trained model to noisy point clouds can produce superior results compared to specialized state-of-the-art techniques. Finally, we demonstrate the utility of our approach in the context of shape reconstruction, by showing how it can be used to extract normal orientation information from point clouds. CCS Concepts •Computing methodologies → Point-based models; Shape analysis; •Computer systems organization → Neural networks; This is a pre-print, the final version is available in the Eurographics 2018 proceedings.

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.

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.

S3I-PointHop: SO(3)-Invariant PointHop for 3D Point Cloud Classification

arXiv (Cornell University), 2023

Many point cloud classification methods are developed under the assumption that all point clouds in the dataset are well aligned with the canonical axes so that the 3D Cartesian point coordinates can be employed to learn features. When input point clouds are not aligned, the classification performance drops significantly. In this work, we focus on a mathematically transparent point cloud classification method called PointHop, analyze its reason for failure due to pose variations, and solve the problem by replacing its pose dependent modules with rotation invariant counterparts. The proposed method is named SO(3)-Invariant PointHop (or S3I-PointHop in short). We also significantly simplify the PointHop pipeline using only one single hop along with multiple spatial aggregation techniques. The idea of exploiting more spatial information is novel. Experiments on the ModelNet40 dataset demonstrate the superiority of S3I-PointHop over traditional PointHop-like methods.