Millions Of Points Research Articles

Optimized quantization parameter selection for video-based point cloud compression

Point clouds are sets of points used to visualize three-dimensional (3D) objects. Point clouds can be static or dynamic. Each point is characterized by its 3D geometry coordinates and attributes such as color. High-quality visualizations often require millions of points, resulting in large storage and transmission costs, especially for dynamic point clouds. To address this problem, the moving picture experts group has recently developed a compression standard for dynamic point clouds called video-based point cloud compression (V-PCC). The standard generates two-dimensional videos from the geometry and color information of the point cloud sequence. Each video is then compressed with a video coder, which converts each frame into frequency coefficients and quantizes them using a quantization parameter (QP). Traditionally, the QPs are severely constrained. For example, in the low-delay configuration of the V-PCC reference software, the quantization parameter values of all the frames in a group of pictures are set to be equal. We show that the rate-distortion performance can be improved by relaxing this constraint and treating the QP selection problem as a multi-variable constrained combinatorial optimization problem, where the variables are the QPs. To solve the optimization problem, we propose a variant of the differential evolution (DE) algorithm. Differential evolution is an evolutionary algorithm that has been successfully applied to various optimization problems. In DE, an initial population of randomly generated candidate solutions is iteratively improved. At each iteration, mutants are generated from the population. Crossover between a mutant and a parent produces offspring. If the performance of the offspring is better than that of the parent, the offspring replaces the parent. While DE was initially introduced for continuous unconstrained optimization problems, we adapt it for our constrained combinatorial optimization problem. Also, unlike standard DE, we apply individual mutation to each variable. Furthermore, we use a variable crossover rate to balance exploration and exploitation. Experimental results for the low-delay configuration of the V-PCC reference software show that our method can reduce the average bitrate by up to 43% compared to a method that uses the same QP values for all frames and selects them according to an interior point method.

Application of Machine Learning to Detect Building Points in Photogrammetry-based Point Clouds

Different point cloud technologies such as Terrestrial Laser Scanners (TLS), Airborne Laser Scanners (ALS), Mobile Mapping Systems (MMS), and Unmanned Aerial Vehicles (UAV) have become increasingly more common in land surveying and geoinformatics over recent years. Thanks to these modern tools, experts can survey large areas cost-effectively with either high resolution or high accuracy. However, processing the point cloud, which consists of millions of points, can be a massive challenge. Manual processing of these large datasets can often be very time-consuming and hardware-demanding, and most of the time, only a limited part of the point cloud is used to derive the final products. The solution can be to automate the process as much as possible. Several advanced mathematical methods, especially Machine Learning (ML) algorithms, allow efficient automated processing of point clouds. This paper presents a processing chain to detect and separate building points from large-scale photogrammetry-based point clouds. The processing is based on the combination of Random Sample Consensus (RANSAC) and Machine Learning (ML) algorithms like Density-Based Spatial Clustering of Applications with Noise (DBSCAN) and Multi-Layer Perceptron (MLP). Presented methods were trained and tested on established and open available Heissigheim 3D (H3D) dataset to separate roof and vegetation points with over 90% accuracy in order to enhance the separation of building points on large-scale point clouds.

Multi-scale hash encoding based neural geometry representation

Recently, neural implicit function-based representation has attracted more and more attention, and has been widely used to represent surfaces using differentiable neural networks. However, surface reconstruction from point clouds or multi-view images using existing neural geometry representations still suffer from slow computation and poor accuracy. To alleviate these issues, we propose a multi-scale hash encoding-based neural geometry representation which effectively and efficiently represents the surface as a signed distance field. Our novel neural network structure carefully combines low-frequency Fourier position encoding with multi-scale hash encoding. The initialization of the geometry network and geometry features of the rendering module are accordingly redesigned. Our experiments demonstrate that the proposed representation is at least 10 times faster for reconstructing point clouds with millions of points. It also significantly improves speed and accuracy of multi-view reconstruction. Our code and models are available at https://github.com/Dengzhi-USTC/Neural-Geometry-Reconstruction.

Point cloud geometry compression with sparse cascaded residuals and sparse attention

AbstractEffective compression of point clouds is essential for implementing virtual and mixed reality applications, which require encoding millions or even tens of millions of points. This paper offers a new geometric compression for point clouds based on sparse cascaded residuals and sparse attention. A sparse cascaded residual module is posited to connect multiple residual modules through shortcuts, thereby augmenting the network's learning capacity and compression efficacy. The authors developed a sparse attention module to acquire global features by computing interdependencies among points, enhancing compression performance to a greater extent. Trade‐off parameters are employed to optimize the rate and distortion. The authors’ method outperforms the state‐of‐the‐art open‐source method regarding rate‐distortion on the ShapeNet, ModelNet, and Microsoft Voxelized Upper Bodies datasets, with average bjøntegaard‐delta (BD)‐rate gains of −14.44% and −15.38%.

VxH: A Systematic Determination of Efficient Hierarchical Voxel Structures

Three-dimensional (3D) maps with many millions to billions of points are now used in an increasing number of applications, with processing rates in the hundreds of thousands to millions of points per second. In mobile applications, power and energy consumption for managing such data and extracting useful information thereof are critical concerns. We have developed structures and methodologies with the purpose of minimizing memory usage and associated energy consumption for indexing and serialization of voxelized point-clouds. The primary source of points in our case is airborne laser scanning, but our methodology is not restricted to only this setting. Our emulated results show a memory usage reduction factor of roughly up to 200× that of Octree/Octomap and a file size reduction factor of up to 1.65× compared with the predominating compression scheme for airborne Lidar data, LASzip. In addition, our structures enable significantly more efficient processing since they are included in a hierarchical structure that captures geometric aspects.

HBIM: A Tool for Enhancing the Diagnosis of Historical Buildings: The Case of St. George’s Memorial Anglican Church, Oshawa

The primary objective of this research is to address the research gap in the conservation of heritage buildings in Canada by integrating Historical Building Information Modeling (HBIM) as a tool. The proposed study aims to develop an enhanced framework for the preservation of historical buildings through the utilization of HBIM and 3D-scanning technology. As a result, the research aims to generate a comprehensive database comprising various families of models while also incorporating strategies for point-cloud clustering data. The significance of this research lies in its potential to contribute to the conservation and restoration process of historical buildings. Currently, there are a lack of standardized approaches and comprehensive databases for accurately documenting and reproducing historical buildings. By integrating HBIM and 3D-scanning technology, this research will enable the creation of highly accurate three-dimensional virtual models, consisting of millions of points, which will serve as a comprehensive dataset for the restoration of heritage buildings. The findings of this research will benefit multiple stakeholders. Preservation architects, conservationists, and heritage professionals will gain a valuable tool for documenting and analyzing historical buildings with a high level of precision. The comprehensive database and framework proposed in this study will facilitate decision-making processes during the restoration and preservation phases, ensuring that the original architectural elements and materials are faithfully reproduced. Additionally, policymakers and governmental organizations involved in heritage conservation can use the outcomes of this research to establish standardized guidelines and regulations for the preservation of historical buildings in Canada. Ultimately, the broader community will benefit from the enhanced preservation efforts, as it will contribute to the cultural and historical identity of the nation, fostering a sense of pride and connection to the past.

Online Stochastic Variational Gaussian Process Mapping for Large-Scale Bathymetric SLAM in Real Time

Rao-Blackwellized particle filter (RBPF) SLAM solutions with Gaussian Process (GP) maps can both maintain multiple hypotheses of a vehicle pose estimate and perform implicit data association for loop closure detection in continuous terrain representations. Both qualities are of particular interest for SLAM with autonomous underwater vehicles (AUVs) in the open sea, where distinguishable features are scarce. However, the applicability of GP regression to parallel, real-time mapping in an RBPF framework remains limited by the size of the area to survey and the computational cost of the GP training. To overcome these constraints, in this letter we propose the adaption of Stochastic Variational GP (SVGP) regression to online mapping in combination with a novel, efficient particle trajectory storing in the RBPF. We show how the resulting RBPF-SVGP framework can achieve real-time performance in an embedded platform on two AUV surveys containing millions of points. We further test the framework on a live mission on an AUV and we make the implementation publicly available.

Unbreaking Assemblies in Molecular Simulations with Periodic Boundaries.

In molecular simulations, periodic boundary conditions are typically used to avoid surface effects occurring at the boundaries of the simulation box. A consequence of this is that molecules and assemblies may appear split over the boundaries. Broken molecular assemblies make it difficult to interpret, analyze, and visualize molecular simulation data. We present a general and fast algorithm that repairs molecular assemblies that are broken due to periodic boundary conditions. The open source method presented here, MDVWhole, works for all translation-only crystallographic periodic boundary conditions. The method consumes little memory and can fix the visualization of the assembly of millions of particles in a few seconds. Thus, it is suitable for processing both single simulation frames and long trajectories with millions of points.

MONITORING OF LAND SURFACE DISPLACEMENT BASED ON SBAS-INSAR TIME-SERIES AND GIS TECHNIQUES: A CASE STUDY OVER THE SHIRAZ METROPOLIS, IRAN

Abstract. Nowadays, subsidence as one of the natural disasters, threaten urban environments. Over the past three decades, the small baseline subset (SBAS) InSAR method due to its capability to monitor medium to large-scale deformations, has emerged as an important tool for estimating surface displacements. In densely vegetated environments, however, the SBAS technique is not effective due to the decorrelation effect. Since the SBAS technique can estimate millions of points for subsidence, these data can be used as input to the Kriging method to estimate the unknown points. Therefore, in this study, we identified the subsidence disaster-stricken-urban areas in the Shiraz metropolis over a 4-year period (between 2017 and 2021) based on the fusion of the SBAS technique and geospatial information science (GIS) using Sentinel-1A time-series images. In the GIS part of this research, the ordinary kriging method was used for interpolation of the SBAS data. Also, the kernel estimation density (KED) was used for the preparation of the raster layers of a population and the built-up density. Based on 33 radar images in ascending and descending orbit, a total of 95 interferograms have been computed. In this research, temporal baselines of less than 100 meters and multi-look window dimensions of 200 meters were used. What is more, the amount of coherence was 0.8 in estimating displacements. Based on the results, the highest displacements occurred in districts 2, 4, 7, and 11. The maximum value of the displacement (30 mm per year) was seen in the southeast of the Shiraz metropolis, in district 7. Moreover, the surface displacement was mainly observed in the lands of the Shiraz international airport and some of the agricultural lands located in district 2. The reasons for the displacements in districts 4 and 11 are related to the high density of the building coverages in these regions. Also, the increased population density was another reason for the subsidence in district 4. It can be concluded that the suggested framework is efficient in determining time-series surface movements in extensive urban and non-urban areas.

Cascaded Contextual Reasoning for Large-Scale Point Cloud Semantic Segmentation

Large-scale point cloud semantic segmentation aims to efficiently process millions of points and classify each point into the correct class. Most traditional methods are designed for indoor scenes and are inpractical to process large-scale point clouds directly. To improve efficiency, appropriate point reduction strategies such as random sampling is necessary to reduce points significantly. However, it brings some problems. First, such reduction leads to serious information loss during encoding phase. Second, to recover from the reduction of points, low-level features need to be carefully fused after upsampling for refinement. To address above issues, we propose a Cascaded Contextual Reasoning network (PointCCR) for large-scale point cloud semantic segmentation. We propose a Dilated Graph Convolution (DGC) module to exploit local structure of point clouds and efficiently increase receptive fields. To make compensation for the information loss caused by reducing points, we propose a Cross-level Feature Enrichment (CFE) module which models cross-level contextual dependencies. DGC and CFE together constitute the Multi-level Cascade Aggregation (MCA) module as encoder. Furthermore, we build a Class-aware Attentive Refinement (CAR) module to selectively fuse low-level features for refinement after upsampling. Extensive experimental results demonstrate our method achieves superior performance compared with other state-of-the-arts on three large-scale point cloud datasets (i.e., Semantic3D, SemanticKITTI and S3DIS).

BLOKÇEYN ŞӘBӘKӘSİNİN KOMPONENTLӘRİNİN İŞLӘNMӘSİ İLӘ İNFORMASİYA MÜBADİLӘSİNİN TӘHLÜKӘSİZLİYİ

Blockchain technology is an emerging technology that has the potential to change the way information is stored and transmitted. Its main components are the concepts of decentralization, transparency and distribution, which, in short, are the absence of any central control system between the source and the destination, the interdependence of the blocks used by the participating parties as a chain, and continuously, the registration and storage, and finally, authorization processes after authentication of the information it needs by means of a key belonging to the original owner, as in asymmetric encryption, with the possibility of any loss and at the time of need, stored at millions of points. The purpose of this summary is to discuss the application of information security for data transmission against cyber attacks in blockchain technology. Different methods of implementing security measures such as encryption, digital signatures and multi-signature transactions, different types of cyber-attacks that blockchain technology is vulnerable to and how to prevent these attacks has also been discussed. Also, the role of smart contracts in ensuring secure data transfer and the importance of developing secure protocols has been emphasized. In terms of security, it has also considered the future of blockchain technology and how it can be used to further increase the security of data transmission.

Isolated Points Prediction via Deep Neural Network on Point Cloud Lossless Geometry Compression

The point cloud is one of the indispensable data structures of virtual and mixed reality applications. Vivid scene requirement means that millions of points need to be encoded while transmitting these point clouds. Therefore, a highly efficient point cloud compression (PCC) technology is the key to promote the above applications. In the tree structure geometry partition based PCC framework, nodes of the tree containing isolated points are further subdivided. This sub-division decreases the compression ratio. Furthermore, the location uncertainty of the isolated points makes the above coding very challenging. In this paper, we propose a deep neural network model to predict the isolated points. These point clouds are generated by LiDAR in the autonomous driving applications. The proposed neural network model includes nodes feature extraction modules and parent-child nodes feature aggregation modules. The nodes feature extraction modules extract the hierarchical features from the inputs generated from the current node or its ancestor nodes. The parent-child nodes feature aggregation modules aggregate the hierarchical feature maps of the parent node and the current node. Finally, this proposed neural network model suggests the probability that the current node contains the isolated points. As a result, the nodes containing isolated points will not be further subdivided. The experimental results show that, compared with the latest isolated points prediction method, the accuracy of the proposed isolated points prediction neural network model is much higher on average. Compared with the state-of-the-art, the proposed model can achieve 5.29% bpip coding gain on average on lossless geometry point cloud compression.

Rethinking Design and Evaluation of 3D Point Cloud Segmentation Models

Currently, the use of 3D point clouds is rapidly increasing in many engineering fields, such as geoscience and manufacturing. Various studies have developed intelligent segmentation models providing accurate results, while only a few of them provide additional insights into the efficiency and robustness of their proposed models. The process of segmentation in the image domain has been studied to a great extent and the research findings are tremendous. However, the segmentation analysis with point clouds is considered particularly challenging due to their unordered and irregular nature. Additionally, solving downstream tasks with 3D point clouds is computationally inefficient, as point clouds normally consist of thousands or millions of points sparsely distributed in 3D space. Thus, there is a significant need for rigorous evaluation of the design characteristics of segmentation models, to be effective and practical. Consequently, in this paper, an in-depth analysis of five fundamental and representative deep learning models for 3D point cloud segmentation is presented. Specifically, we investigate multiple experimental dimensions, such as accuracy, efficiency, and robustness in part segmentation (ShapeNet) and scene segmentation (S3DIS), to assess the effective utilization of the models. Moreover, we create a correspondence between their design properties and experimental properties. For example, we show that convolution-based models that incorporate adaptive weight or position pooling local aggregation operations achieve superior accuracy and robustness to point-wise MLPs, while the latter ones show higher efficiency in time and memory allocation. Our findings pave the way for an effective 3D point cloud segmentation model selection and enlighten the research on point clouds and deep learning.

Investigating the Bioactive Conformation of Angiotensin II Using Markov State Modeling Revisited with Web-Scale Clustering.

Molecular dynamics simulation is a powerful technique for studying the structure and dynamics of biomolecules in atomic-level detail by sampling their various conformations in real time. Because of the long timescales that need to be sampled to study biomolecular processes and the big and complex nature of the corresponding data, relevant analyses of important biophysical phenomena are challenging. Clustering and Markov state models (MSMs) are efficient computational techniques that can be used to extract dominant conformational states and to connect those with kinetic information. In this work, we perform Molecular Dynamics simulations to investigate the free energy landscape of Angiotensin II (AngII) in order to unravel its bioactive conformations using different clustering techniques and Markov state modeling. AngII is an octapeptide hormone, which binds to the AT1 transmembrane receptor, and plays a vital role in the regulation of blood pressure, conservation of total blood volume, and salt homeostasis. To mimic the water-membrane interface as AngII approaches the AT1 receptor and to compare our findings with available experimental results, the simulations were performed in water as well as in water-ethanol mixtures. Our results show that in the water-ethanol environment, AngII adopts more compact U-shaped (folded) conformations than in water, which resembles its structure when bound to the AT1 receptor. For clustering of the conformations, we validate the efficiency of an inverted-quantized k-means algorithm, as a fast approximate clustering technique for web-scale data (millions of points into thousands or millions of clusters) compared to k-means, on data from trajectories of molecular dynamics simulations with reasonable trade-offs between time and accuracy. Finally, we extract MSMs using various clustering techniques for the generation of microstates and macrostates and for the selection of the macrostate representatives.

Utilization of Terrestrial Laser Scanner Technology for Analyzing Shape of Dutch Cave (Bandung, West Java)

Technological developments in the field of surveying and mapping remain in progress. One technology that provides precise and accurate results is the terrestrial laser scanner (TLS). This technology can generate millions of points in a short time and represent objects in 3D. TLS technology has become famous for heritage documentation purposes due to its ability to obtain detailed geometry and information from maintaining historic buildings. This research is in the Dutch Cave, Djuanda Forest Park, Bandung, West Java, to map and document 3D. To produce accurate data, frame measurements were carried out using a total station, GNSS, which ended with acquisition using TLS. The registration method used is the target-to-target method. The goal of this research is to create as-built drawings of the Dutch Cave in analyzing the Dutch Cave's form in greater detail. This investigation determined that the Dutch Cave measures 124.435 meters in length and has a height difference of 69.7 cm between the entrance and the exit. The as-built drawing of the Dutch Cave illustrates that the shape of the cave varies from entrance to exit. There is a cave chamber that widens by 18.8 cm toward the cave's center and narrows by 20.7--32 cm toward the cave's exit. The Dutch Cave has a slope of between 1 and 5 degrees. The Dutch Cave 3D data can be used for a documentation archive for the Djuanda Forest Park.

Fast and Accurate Registration of Terrestrial Point Clouds Using a Planar Approximation of Roof Features

3D reconstructed models are becoming more diffused daily, especially in the Cultural Heritage field. These geometric models are typically obtained from elaborating a 3D point cloud. A significant limit in using these methods is the realignment of different point clouds acquired from different acquisitions, particularly for those whose dimensions are millions of points. Although several methodologies have tried to propose a solution for this necessity, none of these seems to solve definitively the problems related to the realignment of large point clouds. This paper presents a new and innovative procedure for the fine registration of large point clouds. The method performs an alignment by using planar approximations of roof features, taking the roof’s extension into account. It looks particularly suitable for the alignment of large point clouds acquired in urban and archaeological environments. The proposed methodology is compared in terms of accuracy and time with a standard photogrammetric reconstruction based on Ground Control Points (GCPs) and other ones, aligned by the Iterative Closest Point method (ICP) and markers. The results evidence the excellent performance of the methodology, which could represent an alternative for aligning extensive photogrammetric reconstructions without the use of GCPs.

A METHOD FOR EXTRACTING DEFORMATION FEATURES FROM TERRESTRIAL LASER SCANNER 3D POINT CLOUDS DATA IN RGIPT BUILDING

Abstract. With the advancement of measuring technology and the incorporation of photogrammetry into architectural applications, architectural documentation applications have taken on a new viewpoint. Terrestrial laser scanning is a modern approach employed in documentation research today in this field. The following are the most significant advantages of terrestrial laser scanning in this study.The point cloud data obtained by terrestrial laser scanners (TLS) allows for quick access to the correct data at the specified frequency, generating relevant and practical findings for the targeted study, and the ability to use scanners in a variety of working environments. Laser scanners have become one of the most popular tools for obtaining effective and successful outcomes in architectural documentation projects such as surveys, restitution, and renovation. Standardized building rules and design procedures have been developed to create structures that are safe for public usage. Structures are frequently subjected to severe loading situations and difficult environmental situations that were not anticipated during design, resulting in a long-term structural deformation. In this paper, the “RGIPT”, which is one of the educational institutes in Uttar Pradesh, India, was scanned with a terrestrial laser scanner for this study, and 3D sketching, and modelling were done with the SCENE software utilizing only 3D point cloud data. Few columns of the structure have been considered for this paper and their behavior with vibration loading is observed. To conduct this study variation of variances for X-direction and Y-direction is observed for millions of points with increasing height of the structure to predict the upcoming deformation in studied columns of the structure.

Comparison of BIM and Conventional Method to Analyze Dimension and Volume of Abutment

The development of the industrial era 4.0 is rapid, one of which is the BIM (Building Information Modeling) method for surveying activities, namely 3D Scanning. The 3D Scanning method, which can obtain millions of points in point clouds in one scan, is expected to provide higher accuracy and more efficient and effective operating time than conventional methods. Conventional methods, which still apply human plotting, require more time and human resources because they have to move from one point to another, and their accuracy is highly dependent on the quality of human resources. This research was conducted at the Interchange Project of the Batang Industrial Estate, Central Java, and aimed to compare the dimensions, volume, and shop drawings of the 3D Scanning method with the conventional method and the advantages of the 3D Scanning and Modeling method. The results obtained in this study are that there is an average dimension difference of 0.007 m from the shop drawing image, then there is a difference in the average coordinate point of 0.028 m which causes the Abutment position to shift from the shop drawing image. BIM is one method for get the volume of structures on Interchange Project with careful results and faster process. The calculated volume is based on shop drawings. The results showed that the difference between the conventional BIM method in the volume of concrete was 1.156%, and in the granular heap, there was a difference of 1.292%. The advantages of these 2 BIM methods are how they operate, which has started to implement an automation system and requires less time than conventional methods.

High-resolution peak ground acceleration modeling using geographic information systems: A case study of the potentially active Central Cebu Fault System, Philippines

The heavily populated Cebu Island is cut by the Central Cebu Fault System (CCFS). While the CCFS has not produced any MW > 5.0 earthquakes in the past century, recent strong earthquakes in the adjacent islands have brought attention to the seismic hazards in the region. Fault properties such as strike, dip, slip direction, and surface trace length were determined based on literature review, fieldwork, and analysis of geomorphic features. Empirical relations were utilized to estimate down-dip width and magnitude. The gathered data were used to create a three-dimensional model of the four major faults in the CCFS: Balamban Fault, Central Highland Fault, Uling-Masaba Fault, and Lutac-Jaclupan Fault. The 3D model was used to generate peak ground acceleration maps of central Cebu, should an earthquake occur along any of the major faults. Site corrections were made based on the seismic velocity of the upper 30 m of the subsurface. The major faults are estimated to be capable of generating MW 6.4 to 7.1 earthquakes. Worst-case scenarios in densely populated areas show 0.40 to 0.70 g of PGA, suggesting the potential for severe damage in central Cebu. We explored the advantages of using raster mathematics in a GIS platform for calculating and presenting ground motion. These advantages include rapid calculations for tens of millions of points, reducing the effects of interpolation artifacts in final map products. This study emphasizes the importance of detailed structural, geological, and geomorphological data in modeling seismic hazards. Further investigations on the seismogenesis of the CCFS segments are recommended.

Semantic segmentation of large-scale point clouds based on dilated nearest neighbors graph

Three-dimensional (3D) semantic segmentation of point clouds is important in many scenarios, such as automatic driving, robotic navigation, while edge computing is indispensable in the devices. Deep learning methods based on point sampling prove to be computation and memory efficient to tackle large-scale point clouds (e.g. millions of points). However, some local features may be abandoned while sampling. In this paper, We present one end-to-end 3D semantic segmentation framework based on dilated nearest neighbor encoding. Instead of down-sampling point cloud directly, we propose a dilated nearest neighbor encoding module to broaden the network’s receptive field to learn more 3D geometric information. Without increase of network parameters, our method is computation and memory efficient for large-scale point clouds. We have evaluated the dilated nearest neighbor encoding in two different networks. The first is the random sampling with local feature aggregation. The second is the Point Transformer. We have evaluated the quality of the semantic segmentation on the benchmark 3D dataset S3DIS, and demonstrate that the proposed dilated nearest neighbor encoding exhibited stable advantages over baseline and competing methods.