Point Correspondences Research Articles

Axes-aligned non-linear optimized PnP algorithm

Pose estimation is an important component of many real-world computer vision systems. Most existing pose estimation algorithms need a large number of point correspondences to accurately determine the pose of an object. Since the number of point correspondences depends on the object’s appearance, lighting and other external conditions, detecting many points may not be feasible. In many real-world applications, the movement of objects is limited, e.g. due to gravity. Hence, detecting objects with only three degrees of freedom is usually sufficient. This allows us to improve the accuracy of pose estimation by changing the underlying equations of the perspective-n-point problem to three variables instead of six. By using the simplified equations, our algorithm is more robust against detection errors with limited point correspondences. In this article, we study three scenarios where such constraints apply. The first one is about parking a vehicle on a specific spot. Here, a stationary camera is detecting the vehicle to assist the driver. The second scenario describes the perspective of a moving camera detecting objects in its environment. This scenario is common for driver assistance systems, autonomous cars or mobile robots. Third, we describe a camera observing objects from a birds-eye view, which occurs in industrial applications. In all three scenarios, observed objects can only move in the ground plane and rotate around the vertical axis. Hence, three degrees of freedom are sufficient to estimate the pose. Experiments with synthetic data and real-world photographs have shown that our algorithm outperforms state-of-the-art pose estimation algorithms. Depending on the scenario, our algorithm is able to achieve 50% better accuracy, while being equally fast.

Robust and correspondence-free point cloud registration: an extended approach with multiple hypotheses evaluation

AbstractPoint cloud registration is a fundamental problem in robotics, critical for tasks such as localization and mapping. Most approaches to this problem use feature-based techniques. However, these approaches have issues when dealing with unstructured environments where meaningful features are difficult to extract. Recently, an innovative global point cloud registration algorithm, PHASER, which does not rely on geometric features or point correspondences, has been introduced. It leverages Fourier transforms to identify the optimal rigid transform that maximizes cross-correlation between source and target point clouds. PHASER can also incorporate additional data channels, like LiDAR intensity, to enhance registration results. Because it does not rely on local features and because of its ability to exploit additional data, PHASER is particularly useful when dealing with very noisy point clouds or with many outliers. For this reason, we propose an extension to PHASER that considers multiple plausible rototranslation hypotheses. Our extended approach outperforms the original PHASER algorithm, especially in challenging scenarios where point clouds are widely separated. We validate its effectiveness on the DARPA SubT, and the Newer College datasets, showcasing its potential for improving registration accuracy in complex environments.

Camera LiDAR calibration: an automatic and accurate method with novel PLE metrics

Abstract Camera-LiDAR calibration is a crucial aspect of perception systems and sensor fusion in various fields, facilitating the fusion of data from cameras and LiDAR sensors for applications such as autonomous vehicles, robotics, and augmented reality. In this paper, we presented a novel multi-modal fusion method that introduces an efficient camera-LiDAR joint calibration technique using a simple checkerboard. Our method estimates the 3D rigid transformation of the camera system with respect to the LiDAR frame by establishing 2D-to-3D point correspondences for geometric calibration. We improved the idea of the traditional PNP (Perspective-N-Point) algorithm by employing point cloud segmentation and clustering to re-match the checkerboard. When compared with the calibration methods in the Autoware and MATLAB calibration toolbox, the reprojection error was improved by 38.13\% and 58.30\% respectively. The average translation error was improved by 78.26\% and 51.61\% respectively, while the average rotation error is improved by 75\% and 60.78\% respectively, which has a great improvement in all three areas. At the same time, we propose a new evaluation metric for joint calibration results based on reprojected images: PLE (Pixel-Level Evaluation) index to reflect the accuracy of the joint calibration of the camera and LiDAR. An automatic calibration software has been developed for the calibration of the camera's intrinsic parameters as well as for the joint calibration of the camera-LIDAR extrinsic parameters. We have extensively validated our algorithm using the Intel RealSense Depth D435 Camera and LeiShen C16-151B LiDAR.

Automatic 3D Teeth Reconstruction From Five Intra-Oral Photos Using Parametric Teeth Model.

Orthodontic treatment is a lengthy process that requires regular in-person dental monitoring, making remote dental monitoring a viable alternative when face-to-face consultation is not possible. In this study, we propose an improved 3D teeth reconstruction framework that automatically restores the shape, arrangement, and dental occlusion of upper and lower teeth from five intra-oral photographs to aid orthodontists in visualizing the condition of patients in virtual consultations. The framework comprises a parametric model that leverages statistical shape modeling to describe the shape and arrangement of teeth, a modified U-net that extracts teeth contours from intra-oral images, and an iterative process that alternates between finding point correspondences and optimizing a compound loss function to fit the parametric teeth model to predicted teeth contours. We perform a five-fold cross-validation on a dataset of 95 orthodontic cases and report an average Chamfer distance of 1.0121 mm2 and an average Dice similarity coefficient of 0.7672 on all the test samples in the cross-validation, demonstrating a significant improvement compared with the previous work. Our teeth reconstruction framework provides a feasible solution for visualizing 3D teeth models in remote orthodontic consultations.

Fast Robust Point Cloud Registration Based on Compatibility Graph and Accelerated Guided Sampling

Point cloud registration is a crucial technique in photogrammetry, remote sensing, etc. A generalized 3D point cloud registration framework has been developed to estimate the optimal rigid transformation between two point clouds using 3D key point correspondences. However, challenges arise due to the uncertainty in 3D key point detection techniques and the similarity of local surface features. These factors often lead to feature descriptors establishing correspondences containing significant outliers. Current point cloud registration algorithms are typically hindered by these outliers, affecting both their efficiency and accuracy. In this paper, we propose a fast and robust point cloud registration method based on a compatibility graph and accelerated guided sampling. By constructing a compatible graph with correspondences, a minimum subset sampling method combining compatible edge sampling and compatible vertex sampling is proposed to reduce the influence of outliers on the estimation of the registration parameters. Additionally, an accelerated guided sampling strategy based on preference scores is presented, which effectively utilizes model parameters generated during the iterative process to guide the sampling toward inliers, thereby enhancing computational efficiency and the probability of estimating optimal parameters. Experiments are carried out on both synthetic and real-world data. The experimental results demonstrate that our proposed algorithm achieves a significant balance between registration accuracy and efficiency compared to state-of-the-art registration algorithms such as RANSIC and GROR. Even with up to 2000 initial correspondences and an outlier ratio of 99%, our algorithm achieves a minimum rotation error of 0.737° and a minimum translation error of 0.0201 m, completing the registration process within 1 s.

Coarse‐to‐fine adjustment for multi‐platform point cloud fusion

AbstractLeveraging multi‐platform laser scanning systems offers a complete solution for 3D modelling of large‐scale urban scenes. However, the spatial inconsistency of point clouds collected by heterogeneous platforms with different viewpoints presents challenges in achieving seamless fusion. To tackle this challenge, this paper proposes a coarse‐to‐fine adjustment for multi‐platform point cloud fusion. First, in the preprocessing stage, the bounding box of each point cloud block is employed to identify potential constraint association. Second, the proposed local optimisation facilitates preliminary pairwise alignment with these potential constraint relationships, and obtaining initial guess for a comprehensive global optimisation. At last, the proposed global optimisation incorporates all the local constraints for tightly coupled optimisation with raw point correspondences. We choose two study areas to conduct experiments. Study area 1 represents a fast road scene with a significant amount of vegetation, while study area 2 represents an urban scene with many buildings. Extensive experimental evaluations indicate the proposed method has increased the accuracy of study area 1 by 50.6% and the accuracy of study area 2 by 44.7%.

QUANTIFYING HIPPOCAMPAL SHAPE ASYMMETRY IN ALZHEIMER'S DISEASE USING OPTIMAL SHAPE CORRESPONDENCES.

Hippocampal atrophy in Alzheimer's disease (AD) is asymmetric and spatially inhomogeneous. While extensive work has been done on volume and shape analysis of atrophy of the hippocampus in AD, less attention has been given to hippocampal asymmetry specifically. Previous studies of hippocampal asymmetry are limited to global volume or shape measures, which don't localize shape asymmetry at the point level. In this paper, we propose to quantify localized shape asymmetry by optimizing point correspondences between left and right hippocampi within a subject, while simultaneously favoring a compact statistical shape model of the entire sample. To account for related variables that have an impact on AD and healthy subject differences, we build linear models with other confounding factors. Our results on the OASIS3 dataset demonstrate that compared to volumetric information, shape asymmetry reveals fine-grained, localized differences that inform us about the hippocampal regions of most significant shape asymmetry in AD patients.

Weakly Supervised Pose Estimation of Surgical Instrument from a Single Endoscopic Image.

Instrument pose estimation is a key demand in computer-aided surgery, and its main challenges lie in two aspects: Firstly, the difficulty of obtaining stable corresponding image feature points due to the instruments' high refraction and complicated background, and secondly, the lack of labeled pose data. This study aims to tackle the pose estimation problem of surgical instruments in the current endoscope system using a single endoscopic image. More specifically, a weakly supervised method based on the instrument's image segmentation contour is proposed, with the effective assistance of synthesized endoscopic images. Our method consists of the following three modules: a segmentation module to automatically detect the instrument in the input image, followed by a point inference module to predict the image locations of the implicit feature points of the instrument, and a point back-propagatable Perspective-n-Point module to estimate the pose from the tentative 2D-3D corresponding points. To alleviate the over-reliance on point correspondence accuracy, the local errors of feature point matching and the global inconsistency of the corresponding contours are simultaneously minimized. Our proposed method is validated with both real and synthetic images in comparison with the current state-of-the-art methods.

DMS: Low-overlap Registration of 3D Point Clouds with Double-layer Multi-scale Star-graph.

Registering 3D point clouds with low overlap is challenging in 3D computer vision, primarily due to difficulties in identifying small overlap regions and removing correspondence outliers. We observe that the neighborhood similarity can be utilized to detect point correspondence, and the consistent neighborhood correspondence can be used as a criterion to detect robust overlapping regions. So that a Double-layer Multi-scale Star-graph (DMS) structure is proposed to detect robust correspondences using two different types of multi-scale star-graphs. The first-layer Multi-scale Neighbor Feature Star-graphs (MNFS) takes each point as the center and its multi-scale nearest neighbors as the leaves. The MNFS enables to establish the initial correspondence candidate set between the two point clouds based on multi-scale neighborhood topology and feature similarity. Subsequently, each pair of corresponding points find their nearest neighbors within the correspondence sets to construct a Multi-scale Matching Star-graphs (MMS) on each side, so the mutual correspondence relationships between the MMS vertices are identified. These identified mutual correspondences are treated as vertices to construct the Multi-scale Correspondence Star-graphs (MCS), that indicate the relationships among the correspondences. We design edge weight and vertex weight criterion in MCS to detect only the robust correspondence set that has strong neighborhood consistency, so as to reject the outliers. Finally, the point cloud registration is conducted based on the detected robust correspondence. The experimental results demonstrate clearly that the proposed DMS method exhibits superior robustness when compared to existing state-of-the-art registration algorithms. The code of this study will be available at https://github.com/HualongCao/DMS.

Average Degree of the Essential Variety

The essential variety is an algebraic subvariety of dimension 5 in real projective space RP8\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\mathbb R\\mathrm P^{8}$$\\end{document} which encodes the relative pose of two calibrated pinhole cameras. The 5-point algorithm in computer vision computes the real points in the intersection of the essential variety with a linear space of codimension 5. The degree of the essential variety is 10, so this intersection consists of 10 complex points in general. We compute the expected number of real intersection points when the linear space is random. We focus on two probability distributions for linear spaces. The first distribution is invariant under the action of the orthogonal group O(9)\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\ extrm{O}(9)$$\\end{document} acting on linear spaces in RP8\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\mathbb R\\mathrm P^{8}$$\\end{document}. In this case, the expected number of real intersection points is equal to 4. The second distribution is motivated from computer vision and is defined by choosing 5 point correspondences in the image planes RP2×RP2\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$\\mathbb R\\mathrm P^2\ imes \\mathbb R\\mathrm P^2$$\\end{document} uniformly at random. A Monte Carlo computation suggests that with high probability the expected value lies in the interval (3.95-0.05,3.95+0.05)\\documentclass[12pt]{minimal} \\usepackage{amsmath} \\usepackage{wasysym} \\usepackage{amsfonts} \\usepackage{amssymb} \\usepackage{amsbsy} \\usepackage{mathrsfs} \\usepackage{upgreek} \\setlength{\\oddsidemargin}{-69pt} \\begin{document}$$(3.95 - 0.05,\\ 3.95 + 0.05)$$\\end{document}.

High-precision teeth reconstruction based on automatic multimodal fusion with CBCT and IOS

In digital orthodontic treatment, the high-precision reconstruction of complete teeth, encompassing both the crown and the actual root, plays a pivotal role. Current mainstream techniques, prioritizing the high resolution of intraoral scanned models (IOS), are confined to using IOS data for orthodontic treatments. However, the lack of root information in the IOS data may lead to complications such as dehiscence. In contrast, Cone Beam Computed Tomography (CBCT) data encompasses comprehensive dental information with roots. Nonetheless, the radiative character of CBCT scans renders patients unsuitable for repeated examinations in a short time. In addition, lower scanning precision of CBCT leads to suboptimal teeth segmentation outcomes, hindering the accurate representation of dental occlusal relationships. Therefore, in order to fully utilize the complementarity between dental multimodal data, we propose a method for high-precision 3D teeth model reconstruction based on IOS and CBCT, which mainly consists of global rigid registration and local nonrigid registration. Specifically, we extract the priori information of dental arch curves for coarse alignment to provide a good initial position for the Iterative Closest Point (ICP) algorithm, and design a conformal parameterization method for a single tooth to effectively obtain the point correspondence between IOS and CBCT crowns. The rough crown of the CBCT will gradually fit towards the IOS through iterative optimization of nonrigid registration. The experimental results show that our method robustly fuses the advantageous features of IOS and CBCT. The 3D teeth model reconstructed by our method contains the high-precision crown of IOS and the real root of CBCT, which can be effectively used in clinical orthodontic treatment.

An open-source, three-dimensional growth model of the mandible

The available reference data for the mandible and mandibular growth consists primarily of two-dimensional linear or angular measurements. The aim of this study was to create the first open-source, three-dimensional statistical shape model of the mandible that spans the complete growth period. Computed tomography scans of 678 mandibles from children and young adults between 0 and 22 years old were included in the model. The mandibles were segmented using a semi-automatic or automatic (artificial intelligence-based) segmentation method. Point correspondence among the samples was achieved by rigid registration, followed by non-rigid registration of a symmetrical template onto each sample. The registration process was validated with adequate results. Principal component analysis was used to gain insight in the variation within the dataset and to investigate age-related changes and sexual dimorphism. The presented growth model is accessible globally and free-of-charge for scientists, physicians and forensic investigators for any kind of purpose deemed suitable. The versatility of the model opens up new possibilities in the fields of oral and maxillofacial surgery, forensic sciences or biological anthropology. In clinical settings, the model may aid diagnostic decision-making, treatment planning and treatment evaluation.

Point cloud registration method for indoor depth sensor acquisition system based on dual graph computation with irregular shape factors

AbstractThe registration performance determines the widespread indoor application of 3D models acquired by depth sensors. Many advanced registration methods lack comprehensive feature aggregation and poor generalization capabilities, which improves the mismatching ratio. Here, a dual graph network is proposed by incorporating irregular shape factors to make point cloud features more expressive. At first, we transform point cloud sets into the stellar graph within the local neighbourhood of each point. The deep feature and shape factor of each point are combined in the directional‐connected irregular projection space. Subsequently, the combined features are modelled as the second graph. By the attention mechanism computation, feature information is continuously aggregated with intra‐graph and inter‐graph. Finally, a loss function is utilized to confirm point correspondence and perform the registration through singular value decomposition. Extensive experiments validate that the proposed point cloud registration method achieves state‐of‐the‐art performance.

Non-rigid point set registration based on Gaussian mixture model with integrated feature divergence

PurposeThis paper aims to propose a robust method for non-rigid point set registration, using the Gaussian mixture model and accommodating non-rigid transformations. The posterior probabilities of the mixture model are determined through the proposed integrated feature divergence.Design/methodology/approachThe method involves an alternating two-step framework, comprising correspondence estimation and subsequent transformation updating. For correspondence estimation, integrated feature divergences including both global and local features, are coupled with deterministic annealing to address the non-convexity problem of registration. For transformation updating, the expectation-maximization iteration scheme is introduced to iteratively refine correspondence and transformation estimation until convergence.FindingsThe experiments confirm that the proposed registration approach exhibits remarkable robustness on deformation, noise, outliers and occlusion for both 2D and 3D point clouds. Furthermore, the proposed method outperforms existing analogous algorithms in terms of time complexity. Application of stabilizing and securing intermodal containers loaded on ships is performed. The results demonstrate that the proposed registration framework exhibits excellent adaptability for real-scan point clouds, and achieves comparatively superior alignments in a shorter time.Originality/valueThe integrated feature divergence, involving both global and local information of points, is proven to be an effective indicator for measuring the reliability of point correspondences. This inclusion prevents premature convergence, resulting in more robust registration results for our proposed method. Simultaneously, the total operating time is reduced due to a lower number of iterations.

An intelligent recommendation system for personalised parametric garment patterns by integrating designer’s knowledge and 3D body measurements

Garment pattern-making is one of the most important parts of the apparel industry. However, traditional pattern-making is an experience-based work, very time-consuming and ignores the body shape difference. This paper proposes a parametric design method for garment pattern based on body dimensions acquired from a body scanner and body features (body feature points and three segmented body part shape classification) identified by designers according to their professional knowledge. By using this method, we construct a men’s shirt pattern recommendation system oriented to personalised fit. The system consists of two databases and three models. The two databases include a relational database (Database I) and a personalised basic pattern (PBP) database (Database II). The Database I is based on manual and three-dimensional (3D) measurements of human bodies by using designer’s knowledge. And Database I is a relational database, which is organised in terms of the relational model of the body part shape and its key body feature dimensions. After a deep analysis of measured data, the irrelevant measured dimensions to human body shape have been excluded by designers and extract representative human body feature dimensions. In addition, the relations between body shapes and previously identified body feature dimensions have been modelled. From the above relational model, we label key feature point positions on the corresponding 3D body model obtained from 3D body scanning and correct the whole 3D human upper body model into the semantically interpretable one. The 3D personalised basic pattern is drawn on the corrected model based on these key feature points. By using three-dimensional to two-dimensional (3D-to-2D) flattening technology, a 2D flatten graph of the 3D personalised basic pattern of the interpretable model is obtained and slightly adjusted to the form suitable for industrial production, i.e., PBP and the PBP database (Database II) is built. In addition, the three models include a basic pattern parametric model (Model I) (characterizing the relations between the basic pattern and its key influencing human dimensions (chest girth and back length)), a regression model (Model II) which enables to infer from basic pattern to PBP for three body parts based on the one-to-one correspondence of key points between the PBPs and the basic patterns and a personalised shirt pattern parametric model (Model III) (characterizing the structural relations between the personalised shirt pattern (PBPshirt) and PBP). The initial input items of the recommendation system are the body dimension constraint parameters, including chest girth, back length and the body feature dimensions used to determine each body part shape as well as three shirt style constraint parameters (slim, regular and loose). By using Model I, the corresponding basic pattern can be generated through the user’s chest girth and back length. Body feature dimensions determine the three body parts’ shapes. Then, Model II is used to generate the PBP for the corresponding body parts shape. Based on the shirt style chosen by the user, Mode III is used to generate the PBPshirt from the PBP. The output of the recommendation system is a fit-oriented PBPshirt. Moreover, if the PBPshirt is unsatisfactory after a virtual try-on, four adjustable parameters (front side-seam dart, back side-seam dart, waist dart and garment bodice length) are designed to adjust the PBPshirt generated by the proposed recommendation system.

Online Calibration of Extrinsic Parameters for Solid-State LIDAR Systems.

This work addresses the challenge of calibrating multiple solid-state LIDAR systems. The study focuses on three different solid-state LIDAR sensors that implement different hardware designs, leading to distinct scanning patterns for each system. Consequently, detecting corresponding points between the point clouds generated by these LIDAR systems-as required for calibration-is a complex task. To overcome this challenge, this paper proposes a method that involves several steps. First, the measurement data are preprocessed to enhance its quality. Next, features are extracted from the acquired point clouds using the Fast Point Feature Histogram method, which categorizes important characteristics of the data. Finally, the extrinsic parameters are computed using the Fast Global Registration technique. The best set of parameters for the pipeline and the calibration success are evaluated using the normalized root mean square error. In a static real-world indoor scenario, a minimum root mean square error of 7 cm was achieved. Importantly, the paper demonstrates that the presented approach is suitable for online use, indicating its potential for real-time applications. By effectively calibrating the solid-state LIDAR systems and establishing point correspondences, this research contributes to the advancement of multi-LIDAR fusion and facilitates accurate perception and mapping in various fields such as autonomous driving, robotics, and environmental monitoring.

Self-Supervised Bird’s Eye View Motion Prediction with Cross-Modality Signals

Learning the dense bird's eye view (BEV) motion flow in a self-supervised manner is an emerging research for robotics and autonomous driving. Current self-supervised methods mainly rely on point correspondences between point clouds, which may introduce the problems of fake flow and inconsistency, hindering the model’s ability to learn accurate and realistic motion. In this paper, we introduce a novel cross-modality self-supervised training framework that effectively addresses these issues by leveraging multi-modality data to obtain supervision signals. We design three innovative supervision signals to preserve the inherent properties of scene motion, including the masked Chamfer distance loss, the piecewise rigidity loss, and the temporal consistency loss. Through extensive experiments, we demonstrate that our proposed self-supervised framework outperforms all previous self-supervision methods for the motion prediction task.

Implicit Modeling of Non-rigid Objects with Cross-Category Signals

Deep implicit functions (DIFs) have emerged as a potent and articulate means of representing 3D shapes. However, methods modeling object categories or non-rigid entities have mainly focused on single-object scenarios. In this work, we propose MODIF, a multi-object deep implicit function that jointly learns the deformation fields and instance-specific latent codes for multiple objects at once. Our emphasis is on non-rigid, non-interpenetrating entities such as organs. To effectively capture the interrelation between these entities and ensure precise, collision-free representations, our approach facilitates signaling between category-specific fields to adequately rectify shapes. We also introduce novel inter-object supervision: an attraction-repulsion loss is formulated to refine contact regions between objects. Our approach is demonstrated on various medical benchmarks, involving modeling different groups of intricate anatomical entities. Experimental results illustrate that our model can proficiently learn the shape representation of each organ and their relations to others, to the point that shapes missing from unseen instances can be consistently recovered by our method. Finally, MODIF can also propagate semantic information throughout the population via accurate point correspondences.

3-D LiDAR Localization Based on Novel Nonlinear Optimization Method for Autonomous Ground Robot

3D Light Detection and Ranging (LiDAR)-based localization for autonomous ground robot in unknown environment is a critical ability, which has been extensively studied. In this paper, we propose a novel nonlinear optimization method to promote the 3D LiDAR localization performance. Firstly, due to that the objective function is the summation of point correspondence matching error items, a novel item quality evaluation criterion is proposed. The quality of each item is directly proportional to the degree of its matching error distance descent during the optimization process. Then, an attention mechanism based on the criterion is proposed and the objective function is iteratively refined by putting different attention weights on the matching error items. Thirdly, in the Gauss-Newton optimization framework for LiDAR localization, we point out that the Hessian matrix is essentially the sum of the Hessian matrices of high-quality and low-quality point correspondence subsets. To increase the dominance of the Hessian matrix derived from the high-quality point correspondences, the Hessian matrix of the estimated high-quality point correspondence subset in the last updating step is added to the current Hessian matrix. The proposed LiDAR localization is extensively evaluated on public and custom datasets. The experimental results demonstrate that the proposed mechanisms can effectively promote the LiDAR localization performance. We will make our source code open to serve as a new baseline for the robotic community.

HPNet: High precision point cloud registration using feature pyramid and hybrid position encoding

Point cloud registration precision degrades when the scene contains symmetric structures or repeated patches. Also, the common operation of continuous downsampling in existing registration methods causes the loss of detailed information and leads to further registration precision degradation. In this paper, we propose a high precision point cloud registration network (HPNet) to learn discriminative features by considering both multi-scale information and position information. We first propose a multi-scale feature extraction module similar to the feature pyramid that allows the extracted features to contain multi-scale information. Then, an information interaction is performed on the features to learn global contextual information by using Transformer with a hybrid position encoding, which takes into account both absolute and relative positions of points. Finally, the features obtained from the information interaction module are directly used to predict the point correspondences. Comprehensive experiments on 3DMatch, 3DLoMatch, ModelNet, and ModelLoNet datasets demonstrate the state-of-the-art performance of the proposed method and the implementation of HPNet in SLAM shows its effectiveness in real application.