Articoli di riviste sul tema "3D point cloud representation"
Segui questo link per vedere altri tipi di pubblicazioni sul tema: 3D point cloud representation.
Cita una fonte nei formati APA, MLA, Chicago, Harvard e in molti altri stili
Vedi i top-50 articoli di riviste per l'attività di ricerca sul tema "3D point cloud representation".
Accanto a ogni fonte nell'elenco di riferimenti c'è un pulsante "Aggiungi alla bibliografia". Premilo e genereremo automaticamente la citazione bibliografica dell'opera scelta nello stile citazionale di cui hai bisogno: APA, MLA, Harvard, Chicago, Vancouver ecc.
Puoi anche scaricare il testo completo della pubblicazione scientifica nel formato .pdf e leggere online l'abstract (il sommario) dell'opera se è presente nei metadati.
Vedi gli articoli di riviste di molte aree scientifiche e compila una bibliografia corretta.
1
Arya, Hemlata, Parul Saxena e Jaimala Jha. "Detection of 3D Object in Point Cloud: Cloud Semantic Segmentation in Lane Marking". International Journal on Recent and Innovation Trends in Computing and Communication 11, n. 10s (7 ottobre 2023): 376–81. http://dx.doi.org/10.17762/ijritcc.v11i10s.7645.
Abstract (sommario):
Managing a city efficiently and effectively is more important than ever as growing population and economic strain put a strain on infrastructure like transportation and public services like keeping urban green areas clean and maintained. For effective administration, knowledge of the urban setting is essential. Both portable and stationary laser scanners generate 3D point clouds that accurately depict the environment. These data points may be used to infer the state of the roads, buildings, trees, and other important elements involved in this decision-making process. Perhaps they would support "smart" or "smarter" cities in general. Unfortunately, the point clouds do not immediately supply this sort of data. It must be eliminated. This extraction is done either by human specialists or by sophisticated computer programmes that can identify objects. Because the point clouds might represent such large locations, relying on specialists to identify the things may be an unproductive use of time (streets or even whole cities). Automatic or nearly automatic discovery and recognition of essential objects is now possible with the help of object identification software. In this research, In this paper, we describe a unique approach to semantic segmentation of point clouds, based on the usage of contextual point representations to take use of both local and global features within the point cloud. We improve the accuracy of the point's representation by performing a single innovative gated fusion on the point and its neighbours, which incorporates the knowledge from both sets of data and enhances the representation of the point. Following this, we offer a new graph point net module that further develops the improved representation by composing and updating each point's representation inside the local point cloud structure using the graph attention block in real time. Finally, we make advantage of the global structure of the point cloud by using spatial- and channel-wise attention techniques to construct the ensuing semantic label for each point.
2
Barnefske, E., e H. Sternberg. "PCCT: A POINT CLOUD CLASSIFICATION TOOL TO CREATE 3D TRAINING DATA TO ADJUST AND DEVELOP 3D CONVNET". ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLII-2/W16 (17 settembre 2019): 35–40. http://dx.doi.org/10.5194/isprs-archives-xlii-2-w16-35-2019.
Abstract (sommario):
<p><strong>Abstract.</strong> 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.</p>
3
Orts-Escolano, Sergio, Jose Garcia-Rodriguez, Miguel Cazorla, Vicente Morell, Jorge Azorin, Marcelo Saval, Alberto Garcia-Garcia e Victor Villena. "Bioinspired point cloud representation: 3D object tracking". Neural Computing and Applications 29, n. 9 (16 settembre 2016): 663–72. http://dx.doi.org/10.1007/s00521-016-2585-0.
4
Rai, A., N. Srivastava, K. Khoshelham e K. Jain. "SEMANTIC ENRICHMENT OF 3D POINT CLOUDS USING 2D IMAGE SEGMENTATION". International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLVIII-1/W2-2023 (14 dicembre 2023): 1659–66. http://dx.doi.org/10.5194/isprs-archives-xlviii-1-w2-2023-1659-2023.
Abstract (sommario):
Abstract. 3D point cloud segmentation is computationally intensive due to the lack of inherent structural information and the unstructured nature of the point cloud data, which hinders the identification and connection of neighboring points. Understanding the structure of the point cloud data plays a crucial role in obtaining a meaningful and accurate representation of the underlying 3D environment. In this paper, we propose an algorithm that builds on existing state-of-the-art techniques of 2D image segmentation and point cloud registration to enrich point clouds with semantic information. DeepLab2 with ResNet50 as backbone architecture trained on the COCO dataset is used for indoor scene semantic segmentation into several classes like wall, floor, ceiling, doors, and windows. Semantic information from 2D images is propagated along with other input data, i.e., RGB images, depth images, and sensor information to generate 3D point clouds with semantic information. Iterative Closest Point (ICP) algorithm is used for the pair-wise registration of consecutive point clouds and finally, optimization is applied using the pose graph optimization on the whole set of point clouds to generate the combined point cloud of the whole scene. 3D point cloud of the whole scene contains pseudo-color information which denotes the semantic class to which each point belongs. The proposed methodology use an off-the-shelf 2D semantic segmentation deep learning model to semantically segment 3D point clouds collected using handheld mobile LiDAR sensor. We demonstrate a comparison of the accuracy achieved compared to a manually segmented point cloud on an in-house dataset as well as a 2D3DS benchmark dataset.
5
Sun, Yichen. "3D point cloud domain generalization via adversarial training". Applied and Computational Engineering 13, n. 1 (23 ottobre 2023): 160–68. http://dx.doi.org/10.54254/2755-2721/13/20230725.
Abstract (sommario):
The purpose of the paper is to tackle the classification problem of 3D point cloud data in domain generalization: how to develop a generalized feature representation for an unseen target domain by utilizing sub-field of numerous seen source domain(s). We present a novel methodology based on both adversarial training to learn a generalized feature representations across subdomains in domain adaptation called 3D-AA. We specifically expand adversarial autoencoders by applying the Maximum Mean Discrepancy (MMD) measure to align the distributions across several subdomains, and then matching the aligned distribution to any given prior distribution via adversarial feature learning. In this manner, the learned 3D feature representation is supposed to be universal to the observed source domains due to the MMD regularization and is expected to generalize well on the target domain due to the addition of the prior distribution. We applied an algorithm to train two different 3D point cloud source domains with our framework. The combination of multiple loss functions on 3D point cloud domain generalization task show that our applied algorithm performs better and learn more generalized features for the target domain than the source-only algorithm which only utilized the MMD measurement.
6
Yang, Zexin, Qin Ye, Jantien Stoter e Liangliang Nan. "Enriching Point Clouds with Implicit Representations for 3D Classification and Segmentation". Remote Sensing 15, n. 1 (22 dicembre 2022): 61. http://dx.doi.org/10.3390/rs15010061.
Abstract (sommario):
Continuous implicit representations can flexibly describe complex 3D geometry and offer excellent potential for 3D point cloud analysis. However, it remains challenging for existing point-based deep learning architectures to leverage the implicit representations due to the discrepancy in data structures between implicit fields and point clouds. In this work, we propose a new point cloud representation by integrating the 3D Cartesian coordinates with the intrinsic geometric information encapsulated in its implicit field. Specifically, we parameterize the continuous unsigned distance field around each point into a low-dimensional feature vector that captures the local geometry. Then we concatenate the 3D Cartesian coordinates of each point with its encoded implicit feature vector as the network input. The proposed method can be plugged into an existing network architecture as a module without trainable weights. We also introduce a novel local canonicalization approach to ensure the transformation-invariance of encoded implicit features. With its local mechanism, our implicit feature encoding module can be applied to not only point clouds of single objects but also those of complex real-world scenes. We have validated the effectiveness of our approach using five well-known point-based deep networks (i.e., PointNet, SuperPoint Graph, RandLA-Net, CurveNet, and Point Structuring Net) on object-level classification and scene-level semantic segmentation tasks. Extensive experiments on both synthetic and real-world datasets have demonstrated the effectiveness of the proposed point representation.
7
Quach, Maurice, Aladine Chetouani, Giuseppe Valenzise e Frederic Dufaux. "A deep perceptual metric for 3D point clouds". Electronic Imaging 2021, n. 9 (18 gennaio 2021): 257–1. http://dx.doi.org/10.2352/issn.2470-1173.2021.9.iqsp-257.
Abstract (sommario):
Point clouds are essential for storage and transmission of 3D content. As they can entail significant volumes of data, point cloud compression is crucial for practical usage. Recently, point cloud geometry compression approaches based on deep neural networks have been explored. In this paper, we evaluate the ability to predict perceptual quality of typical voxel-based loss functions employed to train these networks. We find that the commonly used focal loss and weighted binary cross entropy are poorly correlated with human perception. We thus propose a perceptual loss function for 3D point clouds which outperforms existing loss functions on the ICIP2020 subjective dataset. In addition, we propose a novel truncated distance field voxel grid representation and find that it leads to sparser latent spaces and loss functions that are more correlated with perceived visual quality compared to a binary representation. The source code is available at <uri>https://github.com/mauriceqch/2021_pc_perceptual_loss</uri>.
8
Decker, Kevin T., e Brett J. Borghetti. "Hyperspectral Point Cloud Projection for the Semantic Segmentation of Multimodal Hyperspectral and Lidar Data with Point Convolution-Based Deep Fusion Neural Networks". Applied Sciences 13, n. 14 (14 luglio 2023): 8210. http://dx.doi.org/10.3390/app13148210.
Abstract (sommario):
The fusion of dissimilar data modalities in neural networks presents a significant challenge, particularly in the case of multimodal hyperspectral and lidar data. Hyperspectral data, typically represented as images with potentially hundreds of bands, provide a wealth of spectral information, while lidar data, commonly represented as point clouds with millions of unordered points in 3D space, offer structural information. The complementary nature of these data types presents a unique challenge due to their fundamentally different representations requiring distinct processing methods. In this work, we introduce an alternative hyperspectral data representation in the form of a hyperspectral point cloud (HSPC), which enables ingestion and exploitation with point cloud processing neural network methods. Additionally, we present a composite fusion-style, point convolution-based neural network architecture for the semantic segmentation of HSPC and lidar point cloud data. We investigate the effects of the proposed HSPC representation for both unimodal and multimodal networks ingesting a variety of hyperspectral and lidar data representations. Finally, we compare the performance of these networks against each other and previous approaches. This study paves the way for innovative approaches to multimodal remote sensing data fusion, unlocking new possibilities for enhanced data analysis and interpretation.
9
Li, Shidi, Miaomiao Liu e Christian Walder. "EditVAE: Unsupervised Parts-Aware Controllable 3D Point Cloud Shape Generation". Proceedings of the AAAI Conference on Artificial Intelligence 36, n. 2 (28 giugno 2022): 1386–94. http://dx.doi.org/10.1609/aaai.v36i2.20027.
Abstract (sommario):
This paper tackles the problem of parts-aware point cloud generation. Unlike existing works which require the point cloud to be segmented into parts a priori, our parts-aware editing and generation are performed in an unsupervised manner. We achieve this with a simple modification of the Variational Auto-Encoder which yields a joint model of the point cloud itself along with a schematic representation of it as a combination of shape primitives. In particular, we introduce a latent representation of the point cloud which can be decomposed into a disentangled representation for each part of the shape. These parts are in turn disentangled into both a shape primitive and a point cloud representation, along with a standardising transformation to a canonical coordinate system. The dependencies between our standardising transformations preserve the spatial dependencies between the parts in a manner that allows meaningful parts-aware point cloud generation and shape editing. In addition to the flexibility afforded by our disentangled representation, the inductive bias introduced by our joint modeling approach yields state-of-the-art experimental results on the ShapeNet dataset.
10
Bello, Saifullahi Aminu, Shangshu Yu, Cheng Wang, Jibril Muhmmad Adam e Jonathan Li. "Review: Deep Learning on 3D Point Clouds". Remote Sensing 12, n. 11 (28 maggio 2020): 1729. http://dx.doi.org/10.3390/rs12111729.
Abstract (sommario):
A point cloud is a set of points defined in a 3D metric space. Point clouds have become one of the most significant data formats for 3D representation and are gaining increased popularity as a result of the increased availability of acquisition devices, as well as seeing increased application in areas such as robotics, autonomous driving, and augmented and virtual reality. Deep learning is now the most powerful tool for data processing in computer vision and is becoming the most preferred technique for tasks such as classification, segmentation, and detection. While deep learning techniques are mainly applied to data with a structured grid, the point cloud, on the other hand, is unstructured. The unstructuredness of point clouds makes the use of deep learning for its direct processing very challenging. This paper contains a review of the recent state-of-the-art deep learning techniques, mainly focusing on raw point cloud data. The initial work on deep learning directly with raw point cloud data did not model local regions; therefore, subsequent approaches model local regions through sampling and grouping. More recently, several approaches have been proposed that not only model the local regions but also explore the correlation between points in the local regions. From the survey, we conclude that approaches that model local regions and take into account the correlation between points in the local regions perform better. Contrary to existing reviews, this paper provides a general structure for learning with raw point clouds, and various methods were compared based on the general structure. This work also introduces the popular 3D point cloud benchmark datasets and discusses the application of deep learning in popular 3D vision tasks, including classification, segmentation, and detection.
11
Lin, Yu, Yigong Wang, Yi-Fan Li, Zhuoyi Wang, Yang Gao e Latifur Khan. "Single View Point Cloud Generation via Unified 3D Prototype". Proceedings of the AAAI Conference on Artificial Intelligence 35, n. 3 (18 maggio 2021): 2064–72. http://dx.doi.org/10.1609/aaai.v35i3.16303.
Abstract (sommario):
As 3D point clouds become the representation of choice for multiple vision and graphics applications, such as autonomous driving, robotics, etc., the generation of them by deep neural networks has attracted increasing attention in the research community. Despite the recent success of deep learning models in classification and segmentation, synthesizing point clouds remains challenging, especially from a single image. State-of-the-art (SOTA) approaches can generate a point cloud from a hidden vector, however, they treat 2D and 3D features equally and disregard the rich shape information within the 3D data. In this paper, we address this problem by integrating image features with 3D prototype features. Specifically, we propose to learn a set of 3D prototype features from a real point cloud dataset and dynamically adjust them through the training. These prototypes are then integrated with incoming image features to guide the point cloud generation process. Experimental results show that our proposed method outperforms SOTA methods on single image based 3D reconstruction tasks.
12
Wang, Yang, e Shunping Xiao. "Affinity-Point Graph Convolutional Network for 3D Point Cloud Analysis". Applied Sciences 12, n. 11 (25 maggio 2022): 5328. http://dx.doi.org/10.3390/app12115328.
Abstract (sommario):
Efficient learning of 3D shape representation from point cloud is one of the biggest requirements in 3D computer vision. In recent years, convolutional neural networks have achieved great success in 2D image representation learning. However, unlike images that have a Euclidean structure, 3D point clouds are irregular since the neighbors of each node are inconsistent. Many studies have tried to develop various convolutional graph neural networks to overcome this problem and to achieve great results. Nevertheless, these studies simply took the centroid point and its corresponding neighbors as the graph structure, thus ignoring the structural information. In this paper, an Affinity-Point Graph Convolutional Network (AP-GCN) is proposed to learn the graph structure for each reference point. In this method, the affinity between points is first defined using the feature of each point feature. Then, a graph with affinity information is built. After that, the edge-conditioned convolution is performed between the graph vertices and edges to obtain stronger neighborhood information. Finally, the learned information is used for recognition and segmentation tasks. Comprehensive experiments demonstrate that AP-GCN learned much more reasonable features and achieved significant improvements in 3D computer vision tasks such as object classification and segmentation.
13
Wang, Tiansheng. "PG-Net:3D point cloud completion based on graph convolutional network". Applied and Computational Engineering 13, n. 1 (23 ottobre 2023): 189–98. http://dx.doi.org/10.54254/2755-2721/13/20230731.
Abstract (sommario):
With the advancement of autonomous driving technology, the problem of 3D point cloud completion has become increasingly important. Completing 3D point clouds can improve the accuracy of 3D object detection, which is crucial for the development of autonomous driving and other related fields. In this paper, we propose a new approach for 3D point cloud completion tasks using point cloud representation. We focuses on the point cloud completion problem using Graph Neural Network methods, which are known for their ability to capture topological features. Our approach utilizes key components extraction and learning from the point cloud, to constrain the output of the decoder, and thus enhance the performance of point cloud completion task. Our approach is able to overcome the limitations of traditional methods, such as memory consumption and computational burden, as well as the loss of detailed information caused by quantization operation in some sparse representation based methods. We conduct extensive experiments on several benchmark datasets to evaluate the performance of our approach and compare it to existing methods. Our experimental results demonstrate that our proposed method is competitive, achieving comparable or even better results compared to state-of-the-art models. In particular, we show that our method is able to improve upon the performance of earlier models and achieve results that are comparable to current state-of-the-art models. These results indicate that our approach is a promising solution for 3D point cloud completion tasks.
14
Yang, Xi, Mengqing Cao, Cong Li, Hua Zhao e Dong Yang. "Learning Implicit Neural Representation for Satellite Object Mesh Reconstruction". Remote Sensing 15, n. 17 (24 agosto 2023): 4163. http://dx.doi.org/10.3390/rs15174163.
Abstract (sommario):
Constructing a surface representation from the sparse point cloud of a satellite is an important task for satellite on-orbit services such as satellite docking and maintenance. In related studies on surface reconstruction from point clouds, implicit neural representations have gained popularity in learning-based 3D object reconstruction. When aiming for a satellite with a more complicated geometry and larger intra-class variance, existing implicit approaches cannot perform well. To solve the above contradictions and make effective use of implicit neural representations, we built a NASA3D dataset containing point clouds, watertight meshes, occupancy values, and corresponding points by using the 3D models on NASA’s official website. On the basis of NASA3D, we propose a novel network called GONet for a more detailed reconstruction of satellite grids. By designing an explicit-related implicit neural representation of the Grid Occupancy Field (GOF) and introducing it into GONet, we compensate for the lack of explicit supervision in existing point cloud surface reconstruction approaches. The GOF, together with the occupancy field (OF), serves as the supervised information for neural network learning. Learning the GOF strengthens GONet’s attention to the critical points of the surface extraction algorithm Marching Cubes; thus, it helps improve the reconstructed surface’s accuracy. In addition, GONet uses the same encoder and decoder as ConvONet but designs a novel Adaptive Feature Aggregation (AFA) module to achieve an adaptive fusion of planar and volume features. The insertion of AFA allows for the obtained implicit features to incorporate more geometric and volumetric information. Both visualization and quantitative experimental results demonstrate that our GONet could handle 3D satellite reconstruction work and outperform existing state-of-the-art methods by a significant margin. With a watertight mesh, our GONet achieves 5.507 CD-L1, 0.8821 F-score, and 68.86% IoU, which is equal to gains of 1.377, 0.0466, and 3.59% over the previous methods using NASA3D, respectively.
15
Zhang, Le, Jian Sun e Qiang Zheng. "3D Point Cloud Recognition Based on a Multi-View Convolutional Neural Network". Sensors 18, n. 11 (29 ottobre 2018): 3681. http://dx.doi.org/10.3390/s18113681.
Abstract (sommario):
The recognition of three-dimensional (3D) lidar (light detection and ranging) point clouds remains a significant issue in point cloud processing. Traditional point cloud recognition employs the 3D point clouds from the whole object. Nevertheless, the lidar data is a collection of two-and-a-half-dimensional (2.5D) point clouds (each 2.5D point cloud comes from a single view) obtained by scanning the object within a certain field angle by lidar. To deal with this problem, we initially propose a novel representation which expresses 3D point clouds using 2.5D point clouds from multiple views and then we generate multi-view 2.5D point cloud data based on the Point Cloud Library (PCL). Subsequently, we design an effective recognition model based on a multi-view convolutional neural network. The model directly acts on the raw 2.5D point clouds from all views and learns to get a global feature descriptor by fusing the features from all views by the view fusion network. It has been proved that our approach can achieve an excellent recognition performance without any requirement for three-dimensional reconstruction and the preprocessing of point clouds. In conclusion, this paper can effectively solve the recognition problem of lidar point clouds and provide vital practical value.
16
Fan, Xiangsuo, Dachuan Xiao, Dengsheng Cai e Wentao Ding. "Real Pseudo-Lidar Point Cloud Fusion for 3D Object Detection". Electronics 12, n. 18 (18 settembre 2023): 3920. http://dx.doi.org/10.3390/electronics12183920.
Abstract (sommario):
Three-dimensional object detection technology is an essential component of autonomous driving systems. Existing 3D object detection techniques heavily rely on expensive lidar sensors, leading to increased costs. Recently, the emergence of Pseudo-Lidar point cloud data has addressed this cost issue. However, the current methods for generating Pseudo-Lidar point clouds are relatively crude, resulting in suboptimal detection performance. This paper proposes an improved method to generate more accurate Pseudo-Lidar point clouds. The method first enhances the stereo-matching network to improve the accuracy of Pseudo-Lidar point cloud representation. Secondly, it fuses 16-Line real lidar point cloud data to obtain more precise Real Pseudo-Lidar point cloud data. Our method achieves impressive results in the popular KITTI benchmark. Our algorithm achieves an object detection accuracy of 85.5% within a range of 30 m. Additionally, the detection accuracies for pedestrians and cyclists reach 68.6% and 61.6%, respectively.
17
Liu, Shaolei, Kexue Fu, Manning Wang e Zhijian Song. "Group-in-Group Relation-Based Transformer for 3D Point Cloud Learning". Remote Sensing 14, n. 7 (24 marzo 2022): 1563. http://dx.doi.org/10.3390/rs14071563.
Abstract (sommario):
Deep point cloud neural networks have achieved promising performance in remote sensing applications, and the prevalence of Transformer in natural language processing and computer vision is in stark contrast to underexplored point-based methods. In this paper, we propose an effective transformer-based network for point cloud learning. To better learn global and local information, we propose a group-in-group relation-based transformer architecture to learn the relationships between point groups to model global information and between points within each group to model local semantic information. To further enhance the local feature representation, we propose a Radius Feature Abstraction (RFA) module to extract radius-based density features characterizing the sparsity of local point clouds. Extensive evaluation on public benchmark datasets demonstrate the effectiveness and competitive performance of our proposed method on point cloud classification and part segmentation.
18
Liu, Weiping, Jia Sun, Wanyi Li, Ting Hu e Peng Wang. "Deep Learning on Point Clouds and Its Application: A Survey". Sensors 19, n. 19 (26 settembre 2019): 4188. http://dx.doi.org/10.3390/s19194188.
Abstract (sommario):
Point cloud is a widely used 3D data form, which can be produced by depth sensors, such as Light Detection and Ranging (LIDAR) and RGB-D cameras. Being unordered and irregular, many researchers focused on the feature engineering of the point cloud. Being able to learn complex hierarchical structures, deep learning has achieved great success with images from cameras. Recently, many researchers have adapted it into the applications of the point cloud. In this paper, the recent existing point cloud feature learning methods are classified as point-based and tree-based. The former directly takes the raw point cloud as the input for deep learning. The latter first employs a k-dimensional tree (Kd-tree) structure to represent the point cloud with a regular representation and then feeds these representations into deep learning models. Their advantages and disadvantages are analyzed. The applications related to point cloud feature learning, including 3D object classification, semantic segmentation, and 3D object detection, are introduced, and the datasets and evaluation metrics are also collected. Finally, the future research trend is predicted.
19
Xu, Mutian, Junhao Zhang, Zhipeng Zhou, Mingye Xu, Xiaojuan Qi e Yu Qiao. "Learning Geometry-Disentangled Representation for Complementary Understanding of 3D Object Point Cloud". Proceedings of the AAAI Conference on Artificial Intelligence 35, n. 4 (18 maggio 2021): 3056–64. http://dx.doi.org/10.1609/aaai.v35i4.16414.
Abstract (sommario):
In 2D image processing, some attempts decompose images into high and low frequency components for describing edge and smooth parts respectively. Similarly, the contour and flat area of 3D objects, such as the boundary and seat area of a chair, describe different but also complementary geometries. However, such investigation is lost in previous deep networks that understand point clouds by directly treating all points or local patches equally. To solve this problem, we propose Geometry-Disentangled Attention Network (GDANet). GDANet introduces Geometry-Disentangle Module to dynamically disentangle point clouds into the contour and flat part of 3D objects, respectively denoted by sharp and gentle variation components. Then GDANet exploits Sharp-Gentle Complementary Attention Module that regards the features from sharp and gentle variation components as two holistic representations, and pays different attentions to them while fusing them respectively with original point cloud features. In this way, our method captures and refines the holistic and complementary 3D geometric semantics from two distinct disentangled components to supplement the local information. Extensive experiments on 3D object classification and segmentation benchmarks demonstrate that GDANet achieves the state-of-the-arts with fewer parameters.
20
El Sayed, Abdul Rahman, Abdallah El Chakik, Hassan Alabboud e Adnan Yassine. "An efficient simplification method for point cloud based on salient regions detection". RAIRO - Operations Research 53, n. 2 (aprile 2019): 487–504. http://dx.doi.org/10.1051/ro/2018082.
Abstract (sommario):
Many computer vision approaches for point clouds processing consider 3D simplification as an important preprocessing phase. On the other hand, the big amount of point cloud data that describe a 3D object require excessively a large storage and long processing time. In this paper, we present an efficient simplification method for 3D point clouds using weighted graphs representation that optimizes the point clouds and maintain the characteristics of the initial data. This method detects the features regions that describe the geometry of the surface. These features regions are detected using the saliency degree of vertices. Then, we define features points in each feature region and remove redundant vertices. Finally, we will show the robustness of our methodviadifferent experimental results. Moreover, we will study the stability of our method according to noise.
21
Hairuddin, A., S. Azri, U. Ujang, M. G. Cuétara, G. M. Retortillo e S. Mohd Salleh. "DEVELOPMENT OF 3D CITY MODEL USING VIDEOGRAMMETRY TECHNIQUE". ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLII-4/W16 (1 ottobre 2019): 221–28. http://dx.doi.org/10.5194/isprs-archives-xlii-4-w16-221-2019.
Abstract (sommario):
Abstract. 3D city model is a representation of urban area in digital format that contains building and other information. The current approaches are using photogrammetry and laser scanning to develop 3D city model. However, these techniques are time consuming and quite costly. Besides that, laser scanning and photogrammetry need professional skills and expertise to handle hardware and tools. In this study, videogrammetry is proposed as a technique to develop 3D city model. This technique uses video frame sequences to generate point cloud. Videos are processed using EyesCloud3D by eCapture. EyesCloud3D allows user to upload raw data of video format to generate point clouds. There are five main phases in this study to generate 3D city model which are calibration, video recording, point cloud extraction, 3D modeling and 3D city model representation. In this study, 3D city model with Level of Detail 2 is produced. Simple query is performed from the database to retrieve the attributes of the 3D city model.
22
Atik, Muhammed Enes, e Zaide Duran. "An Efficient Ensemble Deep Learning Approach for Semantic Point Cloud Segmentation Based on 3D Geometric Features and Range Images". Sensors 22, n. 16 (18 agosto 2022): 6210. http://dx.doi.org/10.3390/s22166210.
Abstract (sommario):
Mobile light detection and ranging (LiDAR) sensor point clouds are used in many fields such as road network management, architecture and urban planning, and 3D High Definition (HD) city maps for autonomous vehicles. Semantic segmentation of mobile point clouds is critical for these tasks. In this study, we present a robust and effective deep learning-based point cloud semantic segmentation method. Semantic segmentation is applied to range images produced from point cloud with spherical projection. Irregular 3D mobile point clouds are transformed into regular form by projecting the clouds onto the plane to generate 2D representation of the point cloud. This representation is fed to the proposed network that produces semantic segmentation. The local geometric feature vector is calculated for each point. Optimum parameter experiments were also performed to obtain the best results for semantic segmentation. The proposed technique, called SegUNet3D, is an ensemble approach based on the combination of U-Net and SegNet algorithms. SegUNet3D algorithm has been compared with five different segmentation algorithms on two challenging datasets. SemanticPOSS dataset includes the urban area, whereas RELLIS-3D includes the off-road environment. As a result of the study, it was demonstrated that the proposed approach is superior to other methods in terms of mean Intersection over Union (mIoU) in both datasets. The proposed method was able to improve the mIoU metric by up to 15.9% in the SemanticPOSS dataset and up to 5.4% in the RELLIS-3D dataset.
23
Zhu, Feng, Jieyu Zhao e Zhengyi Cai. "A Contrastive Learning Method for the Visual Representation of 3D Point Clouds". Algorithms 15, n. 3 (8 marzo 2022): 89. http://dx.doi.org/10.3390/a15030089.
Abstract (sommario):
At present, the unsupervised visual representation learning of the point cloud model is mainly based on generative methods, but the generative methods pay too much attention to the details of each point, thus ignoring the learning of semantic information. Therefore, this paper proposes a discriminative method for the contrastive learning of three-dimensional point cloud visual representations, which can effectively learn the visual representation of point cloud models. The self-attention point cloud capsule network is designed as the backbone network, which can effectively extract the features of point cloud data. By compressing the digital capsule layer, the class dependence of features is eliminated, and the generalization ability of the model and the ability of feature queues to store features are improved. Aiming at the equivariance of the capsule network, the Jaccard loss function is constructed, which is conducive to the network distinguishing the characteristics of positive and negative samples, thereby improving the performance of the contrastive learning. The model is pre-trained on the ShapeNetCore data set, and the pre-trained model is used for classification and segmentation tasks. The classification accuracy on the ModelNet40 data set is 0.1% higher than that of the best unsupervised method, PointCapsNet, and when only 10% of the label data is used, the classification accuracy exceeds 80%. The mIoU of part segmentation on the ShapeNet data set is 1.2% higher than the best comparison method, MulUnsupervised. The experimental results of classification and segmentation show that the proposed method has good performance in accuracy. The alignment and uniformity of features are better than the generative method of PointCapsNet, which proves that this method can learn the visual representation of the three-dimensional point cloud model more effectively.
24
Huang, Rui, Xuran Pan, Henry Zheng, Haojun Jiang, Zhifeng Xie, Cheng Wu, Shiji Song e Gao Huang. "Joint representation learning for text and 3D point cloud". Pattern Recognition 147 (marzo 2024): 110086. http://dx.doi.org/10.1016/j.patcog.2023.110086.
25
Laupheimer, D., M. H. Shams Eddin e N. Haala. "ON THE ASSOCIATION OF LIDAR POINT CLOUDS AND TEXTURED MESHES FOR MULTI-MODAL SEMANTIC SEGMENTATION". ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences V-2-2020 (3 agosto 2020): 509–16. http://dx.doi.org/10.5194/isprs-annals-v-2-2020-509-2020.
Abstract (sommario):
Abstract. The semantic segmentation of the huge amount of acquired 3D data has become an important task in recent years. We propose a novel association mechanism that enables information transfer between two 3D representations: point clouds and meshes. The association mechanism can be used in a two-fold manner: (i) feature transfer to stabilize semantic segmentation of one representation with features from the other representation and (ii) label transfer to achieve the semantic annotation of both representations. We claim that point clouds are an intermediate product whereas meshes are a final user product that jointly provides geometrical and textural information. For this reason, we opt for semantic mesh segmentation in the first place. We apply an off-the-shelf PointNet++ to a textured urban triangle mesh as generated from LiDAR and oblique imagery. For each face within a mesh, a feature vector is computed and optionally extended by inherent LiDAR features as provided by the sensor (e.g. intensity). The feature vector extension is accomplished with the proposed association mechanism. By these means, we leverage inherent features from both data representations for the semantic mesh segmentation (multi-modality). We achieve an overall accuracy of 86:40% on the face-level on a dedicated test mesh. Neglecting LiDAR-inherent features in the per-face feature vectors decreases mean intersection over union by ∼2%. Leveraging our association mechanism, we transfer predicted mesh labels to the LiDAR point cloud at a stroke. To this end, we semantically segment the point cloud by implicit usage of geometric and textural mesh features. The semantic point cloud segmentation achieves an overall accuracy close to 84% on the point-level for both feature vector compositions.
26
Zhang, Jingwen, Zikun Zhou, Guangming Lu, Jiandong Tian e Wenjie Pei. "Robust 3D Tracking with Quality-Aware Shape Completion". Proceedings of the AAAI Conference on Artificial Intelligence 38, n. 7 (24 marzo 2024): 7160–68. http://dx.doi.org/10.1609/aaai.v38i7.28544.
Abstract (sommario):
3D single object tracking remains a challenging problem due to the sparsity and incompleteness of the point clouds. Existing algorithms attempt to address the challenges in two strategies. The first strategy is to learn dense geometric features based on the captured sparse point cloud. Nevertheless, it is quite a formidable task since the learned dense geometric features are with high uncertainty for depicting the shape of the target object. The other strategy is to aggregate the sparse geometric features of multiple templates to enrich the shape information, which is a routine solution in 2D tracking. However, aggregating the coarse shape representations can hardly yield a precise shape representation. Different from 2D pixels, 3D points of different frames can be directly fused by coordinate transform, i.e., shape completion. Considering that, we propose to construct a synthetic target representation composed of dense and complete point clouds depicting the target shape precisely by shape completion for robust 3D tracking. Specifically, we design a voxelized 3D tracking framework with shape completion, in which we propose a quality-aware shape completion mechanism to alleviate the adverse effect of noisy historical predictions. It enables us to effectively construct and leverage the synthetic target representation. Besides, we also develop a voxelized relation modeling module and box refinement module to improve tracking performance. Favorable performance against state-of-the-art algorithms on three benchmarks demonstrates the effectiveness and generalization ability of our method.
27
Ma, Wuwei, Xi Yang, Qiufeng Wang, Kaizhu Huang e Xiaowei Huang. "Multi-Scope Feature Extraction for Intracranial Aneurysm 3D Point Cloud Completion". Cells 11, n. 24 (17 dicembre 2022): 4107. http://dx.doi.org/10.3390/cells11244107.
Abstract (sommario):
3D point clouds are gradually becoming more widely used in the medical field, however, they are rarely used for 3D representation of intracranial vessels and aneurysms due to the time-consuming data reconstruction. In this paper, we simulate the incomplete intracranial vessels (including aneurysms) in the actual collection from different angles, then propose Multi-Scope Feature Extraction Network (MSENet) for Intracranial Aneurysm 3D Point Cloud Completion. MSENet adopts a multi-scope feature extraction encoder to extract the global features from the incomplete point cloud. This encoder utilizes different scopes to fuse the neighborhood information for each point fully. Then a folding-based decoder is applied to obtain the complete 3D shape. To enable the decoder to intuitively match the original geometric structure, we engage the original points coordinates input to perform residual linking. Finally, we merge and sample the complete but coarse point cloud from the decoder to obtain the final refined complete 3D point cloud shape. We conduct extensive experiments on both 3D intracranial aneurysm datasets and general 3D vision PCN datasets. The results demonstrate the effectiveness of the proposed method on three evaluation metrics compared to baseline: our model increases the F-score to 0.379 (+21.1%)/0.320 (+7.7%), reduces Chamfer Distance score to 0.998 (−33.8%)/0.974 (−6.4%), and reduces the Earth Mover’s Distance to 2.750 (17.8%)/2.858 (−0.8%).
28
Poux, F., R. Neuville, P. Hallot e R. Billen. "MODEL FOR SEMANTICALLY RICH POINT CLOUD DATA". ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences IV-4/W5 (23 ottobre 2017): 107–15. http://dx.doi.org/10.5194/isprs-annals-iv-4-w5-107-2017.
Abstract (sommario):
This paper proposes an interoperable model for managing high dimensional point clouds while integrating semantics. Point clouds from sensors are a direct source of information physically describing a 3D state of the recorded environment. As such, they are an exhaustive representation of the real world at every scale: 3D reality-based spatial data. Their generation is increasingly fast but processing routines and data models lack of knowledge to reason from information extraction rather than interpretation. The enhanced smart point cloud developed model allows to bring intelligence to point clouds via 3 connected meta-models while linking available knowledge and classification procedures that permits semantic injection. Interoperability drives the model adaptation to potentially many applications through specialized domain ontologies. A first prototype is implemented in Python and PostgreSQL database and allows to combine semantic and spatial concepts for basic hybrid queries on different point clouds.
29
Chen, Shuaijun, Jinxi Wang, Wei Pan, Shang Gao, Meili Wang e Xuequan Lu. "Towards uniform point distribution in feature-preserving point cloud filtering". Computational Visual Media 9, n. 2 (3 gennaio 2023): 249–63. http://dx.doi.org/10.1007/s41095-022-0278-4.
Abstract (sommario):
AbstractWhile a popular representation of 3D data, point clouds may contain noise and need filtering before use. Existing point cloud filtering methods either cannot preserve sharp features or result in uneven point distributions in the filtered output. To address this problem, this paper introduces a point cloud filtering method that considers both point distribution and feature preservation during filtering. The key idea is to incorporate a repulsion term with a data term in energy minimization. The repulsion term is responsible for the point distribution, while the data term aims to approximate the noisy surfaces while preserving geometric features. This method is capable of handling models with fine-scale features and sharp features. Extensive experiments show that our method quickly yields good results with relatively uniform point distribution.
30
Markiewicz, J. S., Ł. Markiewicz e P. Foryś. "THE COMPARISON OF 2D AND 3D DETECTORS FOR TLS DATA REGISTRATION – PRELIMINARY RESULTS". ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLII-2/W9 (31 gennaio 2019): 467–72. http://dx.doi.org/10.5194/isprs-archives-xlii-2-w9-467-2019.
Abstract (sommario):
<p><strong>Abstract.</strong> This paper presents the analysis of possible methods of a terrestrial laser scanning (TLS) data registration using 2D/3D detectors and descriptors. The developed approach, where point clouds are processed in form of panoramic images, orthoimages and 3D data, was described. The accuracy of the registration process was preliminary verified. The two approaches were analysed and compared: the 2D SIFT (Scale-Invariant Feature Transform) detector and descriptor with the rasterized TLS data and the 3D SIFT detector with the 3D FPFH (Fast Point Feature Histograms) descriptor. The feature points were found and preliminary matched using the OpenCV and PCL (Point Cloud Library) libraries. In order to find the best point cloud representation for the registration process, both the percentage and distribution of the correctly detected and matched points were analysed. The materials consisted of the point clouds of two chambers from the Museum of King John III’s Palace in Wilanów. They were acquired using the Z+F 5006h and 5003 TLS scanners. The performed analysis showed that the lowest RMSE values were for the 2D detectors and orthoimages. However, in the case of the point number and distribution, better results were obtained for using the 3D detector.</p>
31
Xu, Ronghua, Yu Chen, Genshe Chen e Erik Blasch. "SAUSA: Securing Access, Usage, and Storage of 3D Point CloudData by a Blockchain-Based Authentication Network". Future Internet 14, n. 12 (28 novembre 2022): 354. http://dx.doi.org/10.3390/fi14120354.
Abstract (sommario):
The rapid development of three-dimensional (3D) acquisition technology based on 3D sensors provides a large volume of data, which are often represented in the form of point clouds. Point cloud representation can preserve the original geometric information along with associated attributes in a 3D space. Therefore, it has been widely adopted in many scene-understanding-related applications such as virtual reality (VR) and autonomous driving. However, the massive amount of point cloud data aggregated from distributed 3D sensors also poses challenges for secure data collection, management, storage, and sharing. Thanks to the characteristics of decentralization and security, Blockchain has great potential to improve point cloud services and enhance security and privacy preservation. Inspired by the rationales behind the software-defined network (SDN) technology, this paper envisions SAUSA, a Blockchain-based authentication network that is capable of recording, tracking, and auditing the access, usage, and storage of 3D point cloud datasets in their life-cycle in a decentralized manner. SAUSA adopts an SDN-inspired point cloud service architecture, which allows for efficient data processing and delivery to satisfy diverse quality-of-service (QoS) requirements. A Blockchain-based authentication framework is proposed to ensure security and privacy preservation in point cloud data acquisition, storage, and analytics. Leveraging smart contracts for digitizing access control policies and point cloud data on the Blockchain, data owners have full control of their 3D sensors and point clouds. In addition, anyone can verify the authenticity and integrity of point clouds in use without relying on a third party. Moreover, SAUSA integrates a decentralized storage platform to store encrypted point clouds while recording references of raw data on the distributed ledger. Such a hybrid on-chain and off-chain storage strategy not only improves robustness and availability, but also ensures privacy preservation for sensitive information in point cloud applications. A proof-of-concept prototype is implemented and tested on a physical network. The experimental evaluation validates the feasibility and effectiveness of the proposed SAUSA solution.
32
Huang, Xiaoshui, Zhou Huang, Sheng Li, Wentao Qu, Tong He, Yuenan Hou, Yifan Zuo e Wanli Ouyang. "Frozen CLIP Transformer Is an Efficient Point Cloud Encoder". Proceedings of the AAAI Conference on Artificial Intelligence 38, n. 3 (24 marzo 2024): 2382–90. http://dx.doi.org/10.1609/aaai.v38i3.28013.
Abstract (sommario):
The pretrain-finetune paradigm has achieved great success in NLP and 2D image fields because of the high-quality representation ability and transferability of their pretrained models. However, pretraining such a strong model is difficult in the 3D point cloud field due to the limited amount of point cloud sequences. This paper introduces Efficient Point Cloud Learning (EPCL), an effective and efficient point cloud learner for directly training high-quality point cloud models with a frozen CLIP transformer. Our EPCL connects the 2D and 3D modalities by semantically aligning the image features and point cloud features without paired 2D-3D data. Specifically, the input point cloud is divided into a series of local patches, which are converted to token embeddings by the designed point cloud tokenizer. These token embeddings are concatenated with a task token and fed into the frozen CLIP transformer to learn point cloud representation. The intuition is that the proposed point cloud tokenizer projects the input point cloud into a unified token space that is similar to the 2D images. Comprehensive experiments on 3D detection, semantic segmentation, classification and few-shot learning demonstrate that the CLIP transformer can serve as an efficient point cloud encoder and our method achieves promising performance on both indoor and outdoor benchmarks. In particular, performance gains brought by our EPCL are 19.7 AP50 on ScanNet V2 detection, 4.4 mIoU on S3DIS segmentation and 1.2 mIoU on SemanticKITTI segmentation compared to contemporary pretrained models. Code is available at \url{https://github.com/XiaoshuiHuang/EPCL}.
33
You, Haoxuan, Yifan Feng, Xibin Zhao, Changqing Zou, Rongrong Ji e Yue Gao. "PVRNet: Point-View Relation Neural Network for 3D Shape Recognition". Proceedings of the AAAI Conference on Artificial Intelligence 33 (17 luglio 2019): 9119–26. http://dx.doi.org/10.1609/aaai.v33i01.33019119.
Abstract (sommario):
Three-dimensional (3D) shape recognition has drawn much research attention in the field of computer vision. The advances of deep learning encourage various deep models for 3D feature representation. For point cloud and multi-view data, two popular 3D data modalities, different models are proposed with remarkable performance. However the relation between point cloud and views has been rarely investigated. In this paper, we introduce Point-View Relation Network (PVRNet), an effective network designed to well fuse the view features and the point cloud feature with a proposed relation score module. More specifically, based on the relation score module, the point-single-view fusion feature is first extracted by fusing the point cloud feature and each single view feature with point-singe-view relation, then the pointmulti- view fusion feature is extracted by fusing the point cloud feature and the features of different number of views with point-multi-view relation. Finally, the point-single-view fusion feature and point-multi-view fusion feature are further combined together to achieve a unified representation for a 3D shape. Our proposed PVRNet has been evaluated on ModelNet40 dataset for 3D shape classification and retrieval. Experimental results indicate our model can achieve significant performance improvement compared with the state-of-the-art models.
34
Liu, Huaijin, Jixiang Du, Yong Zhang e Hongbo Zhang. "Enhancing Point Features with Spatial Information for Point-Based 3D Object Detection". Scientific Programming 2021 (21 dicembre 2021): 1–11. http://dx.doi.org/10.1155/2021/4650660.
Abstract (sommario):
Currently, there are many kinds of voxel-based multisensor 3D object detectors, while point-based multisensor 3D object detectors have not been fully studied. In this paper, we propose a new 3D two-stage object detection method based on point cloud and image fusion to improve the detection accuracy. To address the problem of insufficient semantic information of point cloud, we perform multiscale deep fusion of LiDAR point and camera image in a point-wise manner to enhance point features. Due to the imbalance of LiDAR points, the object point cloud in the long-distance area is sparse. We design a point cloud completion module to predict the spatial shape of objects in the candidate boxes and extract the structural information to improve the feature representation ability to further refine the boxes. The framework is evaluated on widely used KITTI and SUN-RGBD dataset. Experimental results show that our method outperforms all state-of-the-art point-based 3D object detection methods and has comparable performance to voxel-based methods as well.
35
Firintepe, Ahmet, Carolin Vey, Stylianos Asteriadis, Alain Pagani e Didier Stricker. "From IR Images to Point Clouds to Pose: Point Cloud-Based AR Glasses Pose Estimation". Journal of Imaging 7, n. 5 (27 aprile 2021): 80. http://dx.doi.org/10.3390/jimaging7050080.
Abstract (sommario):
In this paper, we propose two novel AR glasses pose estimation algorithms from single infrared images by using 3D point clouds as an intermediate representation. Our first approach “PointsToRotation” is based on a Deep Neural Network alone, whereas our second approach “PointsToPose” is a hybrid model combining Deep Learning and a voting-based mechanism. Our methods utilize a point cloud estimator, which we trained on multi-view infrared images in a semi-supervised manner, generating point clouds based on one image only. We generate a point cloud dataset with our point cloud estimator using the HMDPose dataset, consisting of multi-view infrared images of various AR glasses with the corresponding 6-DoF poses. In comparison to another point cloud-based 6-DoF pose estimation named CloudPose, we achieve an error reduction of around 50%. Compared to a state-of-the-art image-based method, we reduce the pose estimation error by around 96%.
36
Yu, Siyang, Si Sun, Wei Yan, Guangshuai Liu e Xurui Li. "A Method Based on Curvature and Hierarchical Strategy for Dynamic Point Cloud Compression in Augmented and Virtual Reality System". Sensors 22, n. 3 (7 febbraio 2022): 1262. http://dx.doi.org/10.3390/s22031262.
Abstract (sommario):
As a kind of information-intensive 3D representation, point cloud rapidly develops in immersive applications, which has also sparked new attention in point cloud compression. The most popular dynamic methods ignore the characteristics of point clouds and use an exhaustive neighborhood search, which seriously impacts the encoder’s runtime. Therefore, we propose an improved compression means for dynamic point cloud based on curvature estimation and hierarchical strategy to meet the demands in real-world scenarios. This method includes initial segmentation derived from the similarity between normals, curvature-based hierarchical refining process for iterating, and image generation and video compression technology based on de-redundancy without performance loss. The curvature-based hierarchical refining module divides the voxel point cloud into high-curvature points and low-curvature points and optimizes the initial clusters hierarchically. The experimental results show that our method achieved improved compression performance and faster runtime than traditional video-based dynamic point cloud compression.
37
Poliyapram, Vinayaraj, Weimin Wang e Ryosuke Nakamura. "A Point-Wise LiDAR and Image Multimodal Fusion Network (PMNet) for Aerial Point Cloud 3D Semantic Segmentation". Remote Sensing 11, n. 24 (10 dicembre 2019): 2961. http://dx.doi.org/10.3390/rs11242961.
Abstract (sommario):
3D semantic segmentation of point cloud aims at assigning semantic labels to each point by utilizing and respecting the 3D representation of the data. Detailed 3D semantic segmentation of urban areas can assist policymakers, insurance companies, governmental agencies for applications such as urban growth assessment, disaster management, and traffic supervision. The recent proliferation of remote sensing techniques has led to producing high resolution multimodal geospatial data. Nonetheless, currently, only limited technologies are available to fuse the multimodal dataset effectively. Therefore, this paper proposes a novel deep learning-based end-to-end Point-wise LiDAR and Image Multimodal Fusion Network (PMNet) for 3D segmentation of aerial point cloud by fusing aerial image features. PMNet respects basic characteristics of point cloud such as unordered, irregular format and permutation invariance. Notably, multi-view 3D scanned data can also be trained using PMNet since it considers aerial point cloud as a fully 3D representation. The proposed method was applied on two datasets (1) collected from the urban area of Osaka, Japan and (2) from the University of Houston campus, USA and its neighborhood. The quantitative and qualitative evaluation shows that PMNet outperforms other models which use non-fusion and multimodal fusion (observational-level fusion and feature-level fusion) strategies. In addition, the paper demonstrates the improved performance of the proposed model (PMNet) by over-sampling/augmenting the medium and minor classes in order to address the class-imbalance issues.
38
Hoang, Long, Suk-Hwan Lee, Eung-Joo Lee e Ki-Ryong Kwon. "GSV-NET: A Multi-Modal Deep Learning Network for 3D Point Cloud Classification". Applied Sciences 12, n. 1 (4 gennaio 2022): 483. http://dx.doi.org/10.3390/app12010483.
Abstract (sommario):
Light Detection and Ranging (LiDAR), which applies light in the formation of a pulsed laser to estimate the distance between the LiDAR sensor and objects, is an effective remote sensing technology. Many applications use LiDAR including autonomous vehicles, robotics, and virtual and augmented reality (VR/AR). The 3D point cloud classification is now a hot research topic with the evolution of LiDAR technology. This research aims to provide a high performance and compatible real-world data method for 3D point cloud classification. More specifically, we introduce a novel framework for 3D point cloud classification, namely, GSV-NET, which uses Gaussian Supervector and enhancing region representation. GSV-NET extracts and combines both global and regional features of the 3D point cloud to further enhance the information of the point cloud features for the 3D point cloud classification. Firstly, we input the Gaussian Supervector description into a 3D wide-inception convolution neural network (CNN) structure to define the global feature. Secondly, we convert the regions of the 3D point cloud into color representation and capture region features with a 2D wide-inception network. These extracted features are inputs of a 1D CNN architecture. We evaluate the proposed framework on the point cloud dataset: ModelNet and the LiDAR dataset: Sydney. The ModelNet dataset was developed by Princeton University (New Jersey, United States), while the Sydney dataset was created by the University of Sydney (Sydney, Australia). Based on our numerical results, our framework achieves more accuracy than the state-of-the-art approaches.
39
Xu, Mingye, Zhipeng Zhou, Junhao Zhang e Yu Qiao. "Investigate Indistinguishable Points in Semantic Segmentation of 3D Point Cloud". Proceedings of the AAAI Conference on Artificial Intelligence 35, n. 4 (18 maggio 2021): 3047–55. http://dx.doi.org/10.1609/aaai.v35i4.16413.
Abstract (sommario):
This paper investigates the indistinguishable points (difficult to predict label) in semantic segmentation for large-scale 3D point clouds. The indistinguishable points consist of those located in complex boundary, points with similar local textures but different categories, and points in isolate small hard areas, which largely harm the performance of 3D semantic segmentation. To address this challenge, we propose a novel Indistinguishable Area Focalization Network (IAF-Net), which select indistinguishable points adaptively by utilizing the hierarchical semantic features and enhance fine-grained features for points especially those indistinguishable points. We also introduce multi-stage loss to improve the feature representation in a progressive way. Moreover, in order to analyze the segmentation performances of indistinguishable areas, we propose a new evaluation metric called Indistinguishable Points Based Metric (IPBM). Our IAF-Net achieves the state-of-the-art performance on several popular 3D point datasets e.g. S3DIS and ScanNet, and clearly outperform other methods on IPBM. Our code will be available at https://github.com/MingyeXu/IAF-Net.
40
Huang, Ming, Xueyu Wu, Xianglei Liu, Tianhang Meng e Peiyuan Zhu. "Integration of Constructive Solid Geometry and Boundary Representation (CSG-BRep) for 3D Modeling of Underground Cable Wells from Point Clouds". Remote Sensing 12, n. 9 (4 maggio 2020): 1452. http://dx.doi.org/10.3390/rs12091452.
Abstract (sommario):
The preference of three-dimensional representation of underground cable wells from two-dimensional symbols is a developing trend, and three-dimensional (3D) point cloud data is widely used due to its high precision. In this study, we utilize the characteristics of 3D terrestrial lidar point cloud data to build a CSG-BRep 3D model of underground cable wells, whose spatial topological relationship is fully considered. In order to simplify the modeling process, first, point cloud simplification is performed; then, the point cloud main axis is extracted by OBB bounding box, and lastly the point cloud orientation correction is realized by quaternion rotation. Furthermore, employing the adaptive method, the top point cloud is extracted, and it is projected for boundary extraction. Thereupon, utilizing the boundary information, we design the 3D cable well model. Finally, the cable well component model is generated by scanning the original point cloud. The experiments demonstrate that, along with the algorithm being fast, the proposed model is effective at displaying the 3D information of the actual cable wells and meets the current production demands.
41
Leal, Esmeide, German Sanchez-Torres, John W. Branch-Bedoya, Francisco Abad e Nallig Leal. "A Saliency-Based Sparse Representation Method for Point Cloud Simplification". Sensors 21, n. 13 (23 giugno 2021): 4279. http://dx.doi.org/10.3390/s21134279.
Abstract (sommario):
High-resolution 3D scanning devices produce high-density point clouds, which require a large capacity of storage and time-consuming processing algorithms. In order to reduce both needs, it is common to apply surface simplification algorithms as a preprocessing stage. The goal of point cloud simplification algorithms is to reduce the volume of data while preserving the most relevant features of the original point cloud. In this paper, we present a new point cloud feature-preserving simplification algorithm. We use a global approach to detect saliencies on a given point cloud. Our method estimates a feature vector for each point in the cloud. The components of the feature vector are the normal vector coordinates, the point coordinates, and the surface curvature at each point. Feature vectors are used as basis signals to carry out a dictionary learning process, producing a trained dictionary. We perform the corresponding sparse coding process to produce a sparse matrix. To detect the saliencies, the proposed method uses two measures, the first of which takes into account the quantity of nonzero elements in each column vector of the sparse matrix and the second the reconstruction error of each signal. These measures are then combined to produce the final saliency value for each point in the cloud. Next, we proceed with the simplification of the point cloud, guided by the detected saliency and using the saliency values of each point as a dynamic clusterization radius. We validate the proposed method by comparing it with a set of state-of-the-art methods, demonstrating the effectiveness of the simplification method.
42
Razali, A. F., M. F. M. Ariff e Z. Majid. "A HYBRID POINT CLOUD REALITY CAPTURE FROM TERRESTRIAL LASER SCANNING AND UAV-PHOTOGRAMMETRY". International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLVI-2/W1-2022 (25 febbraio 2022): 459–63. http://dx.doi.org/10.5194/isprs-archives-xlvi-2-w1-2022-459-2022.
Abstract (sommario):
Abstract. Point clouds are a digital representation of physical objects or buildings that exist in real world. There are many sources that a point cloud can come from such as a terrestrial laser scanner (TLS) or an unmanned aerial vehicle (UAV). This paper presents a simple method of integrating point clouds from two (2) data sources; TLS and UAV using simple alignment of rigid body transformation method known as Point Pair Picking (PPP). The point cloud data are the representation of details of a one-story building located in Johor Bahru, Malaysia. The process of aligning two (2) separate clouds into one (1) dataset requires initial processing such as noise removal before the alignment process was started. A laser (LAS) formatted data were formed so that it compatible with the PPP process. As the result, a high dense hybrid cloud-model was produced covering complete details of the building. This shows that integration of point clouds could improve 3D documentation assessment such as Building Information Modelling (BIM) by contributing richer semantic information.
43
Imdad, Ulfat, Mirza Tahir Ahmed, Muhammad Asif e Hanan Aljuaid. "3D point cloud lossy compression using quadric surfaces". PeerJ Computer Science 7 (6 ottobre 2021): e675. http://dx.doi.org/10.7717/peerj-cs.675.
Abstract (sommario):
The presence of 3D sensors in hand-held or head-mounted smart devices has motivated many researchers around the globe to devise algorithms to manage 3D point cloud data efficiently and economically. This paper presents a novel lossy compression technique to compress and decompress 3D point cloud data that will save storage space on smart devices as well as minimize the use of bandwidth when transferred over the network. The idea presented in this research exploits geometric information of the scene by using quadric surface representation of the point cloud. A region of a point cloud can be represented by the coefficients of quadric surface when the boundary conditions are known. Thus, a set of quadric surface coefficients and their associated boundary conditions are stored as a compressed point cloud and used to decompress. An added advantage of proposed technique is its flexibility to decompress the cloud as a dense or a course cloud. We compared our technique with state-of-the-art 3D lossless and lossy compression techniques on a number of standard publicly available datasets with varying the structure complexities.
44
Castagno, Jeremy, e Ella Atkins. "Polylidar3D-Fast Polygon Extraction from 3D Data". Sensors 20, n. 17 (26 agosto 2020): 4819. http://dx.doi.org/10.3390/s20174819.
Abstract (sommario):
Flat surfaces captured by 3D point clouds are often used for localization, mapping, and modeling. Dense point cloud processing has high computation and memory costs making low-dimensional representations of flat surfaces such as polygons desirable. We present Polylidar3D, a non-convex polygon extraction algorithm which takes as input unorganized 3D point clouds (e.g., LiDAR data), organized point clouds (e.g., range images), or user-provided meshes. Non-convex polygons represent flat surfaces in an environment with interior cutouts representing obstacles or holes. The Polylidar3D front-end transforms input data into a half-edge triangular mesh. This representation provides a common level of abstraction for subsequent back-end processing. The Polylidar3D back-end is composed of four core algorithms: mesh smoothing, dominant plane normal estimation, planar segment extraction, and finally polygon extraction. Polylidar3D is shown to be quite fast, making use of CPU multi-threading and GPU acceleration when available. We demonstrate Polylidar3D’s versatility and speed with real-world datasets including aerial LiDAR point clouds for rooftop mapping, autonomous driving LiDAR point clouds for road surface detection, and RGBD cameras for indoor floor/wall detection. We also evaluate Polylidar3D on a challenging planar segmentation benchmark dataset. Results consistently show excellent speed and accuracy.
45
Lin, Guoting, Zexun Zheng, Lin Chen, Tianyi Qin e Jiahui Song. "Multi-Modal 3D Shape Clustering with Dual Contrastive Learning". Applied Sciences 12, n. 15 (22 luglio 2022): 7384. http://dx.doi.org/10.3390/app12157384.
Abstract (sommario):
3D shape clustering is developing into an important research subject with the wide applications of 3D shapes in computer vision and multimedia fields. Since 3D shapes generally take on various modalities, how to comprehensively exploit the multi-modal properties to boost clustering performance has become a key issue for the 3D shape clustering task. Taking into account the advantages of multiple views and point clouds, this paper proposes the first multi-modal 3D shape clustering method, named the dual contrastive learning network (DCL-Net), to discover the clustering partitions of unlabeled 3D shapes. First, by simultaneously performing cross-view contrastive learning within multi-view modality and cross-modal contrastive learning between the point cloud and multi-view modalities in the representation space, a representation-level dual contrastive learning module is developed, which aims to capture discriminative 3D shape features for clustering. Meanwhile, an assignment-level dual contrastive learning module is designed by further ensuring the consistency of clustering assignments within the multi-view modality, as well as between the point cloud and multi-view modalities, thus obtaining more compact clustering partitions. Experiments on two commonly used 3D shape benchmarks demonstrate the effectiveness of the proposed DCL-Net.
46
Ryu, Min Woo, Sang Min Oh, Min Ju Kim, Hun Hee Cho, Chang Baek Son e Tae Hoon Kim. "Algorithm for Generating 3D Geometric Representation Based on Indoor Point Cloud Data". Applied Sciences 10, n. 22 (14 novembre 2020): 8073. http://dx.doi.org/10.3390/app10228073.
Abstract (sommario):
This study proposes a new method to generate a three-dimensional (3D) geometric representation of an indoor environment by refining and processing an indoor point cloud data (PCD) captured through backpack laser scanners. The proposed algorithm comprises two parts to generate the 3D geometric representation: data refinement and data processing. In the refinement section, the inputted indoor PCD are roughly segmented by applying random sample consensus (RANSAC) to raw data based on an estimated normal vector. Next, the 3D geometric representation is generated by calculating and separating tangent points on segmented PCD. This study proposes a robust algorithm that utilizes the topological feature of the indoor PCD created by a hierarchical data process. The algorithm minimizes the size and the uncertainty of raw PCD caused by the absence of a global navigation satellite system and equipment errors. The result of this study shows that the indoor environment can be converted into 3D geometric representation by applying the proposed algorithm to the indoor PCD.
47
Stojanovic, V., M. Trapp, R. Richter e J. Döllner. "A SERVICE-ORIENTED INDOOR POINT CLOUD PROCESSING PIPELINE". ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLII-2/W17 (29 novembre 2019): 339–46. http://dx.doi.org/10.5194/isprs-archives-xlii-2-w17-339-2019.
Abstract (sommario):
Abstract. Visualization of point clouds plays an important role in understanding the context of the digital representation of the built environment. Modern commodity mobile devices (e.g., smartphones and tablets), are capable of capturing representations in the form of 3D point clouds, with their depth-sensing and photogrammetry capabilities. Points clouds enable the encoding of important spatial and physical features of the built environment they represent. However, once captured, point clouds need to be processed before they can be used for further semantic enrichment and decision making. An integrated pipeline for such processes is crucial for use in larger and more complex enterprise systems and data analysis platforms, especially within the realm of Facility Management (FM) and Real Estate 4.0. We present and discuss a prototypical implementation for a service-oriented point cloud processing pipeline. The presented processing features focus on detecting and visualizing spatial deviations between as-is versus as-designed representations. We discuss the design and implementation of these processing features, and present experimental results. The presented approach can be used as a lightweight software component for processing indoor point clouds captured using commodity mobile devices, as well as primary deviation analysis, and also provides a processing link for further semantic enrichment of base-data for Building Information Modeling (BIM) and Digital Twin (DT) applications.
48
Beckmann, Sophie, Jean-Claude Rosenthal, Eric L. Wisotzky, Peter Eisert e Anna Hilsmann. "Automatic Registration of Anatomical Structures of Stereo-Endoscopic Point Clouds". Current Directions in Biomedical Engineering 9, n. 1 (1 settembre 2023): 615–18. http://dx.doi.org/10.1515/cdbme-2023-1154.
Abstract (sommario):
Abstract In image-guided surgery, imaging systems such as (stereo-)endoscopes allow intra-operative 3D reconstructions in form of point clouds. However, endoscopes provide only a narrow field of view, resulting in a confined point cloud. In this paper, we present an analysis and registration pipeline for confined point clouds acquired by stereo endoscopes into a fused representation. For a coarse registration, TEASER is applied, while a refinement is conducted utilizing point-to-plane ICP. The pipeline is tested on two datasets: acquired point clouds of a head phantom using a EinsteinVision® 3.0 endoscope and point clouds from the Stereo Correspondence And Reconstruction of Endoscopic Data challenge. The results for both datasets show that 3D reconstructions of anatomical structures by utilizing stereo-endoscopes and point cloud registration is a promising contactless and radiation-free method. However, non-rigid deformations are not yet incorporated and evaluation of the method compared to reference data is challenging.
49
Du, Han, Benhe Cai, Xiaoming Li, Weixi Wang e Shengjun Tang. "Method for Generating Indoor 3D Scene Graphs Based on Instance Features and Relationship Encoding". International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLVIII-1-2024 (10 maggio 2024): 135–40. http://dx.doi.org/10.5194/isprs-archives-xlviii-1-2024-135-2024.
Abstract (sommario):
Abstract. A 3D scene graph is a compact and explicit representation in scene analysis. In today’s 3D scene graph prediction methods, the feature encoding method of nodes and edges is relatively simple, which essentially hinders the network from fully learning 3D point cloud features. In this paper, we propose a 3D scene graph task framework that fully expresses node and edge features, trying to meet the requirements of fully utilizing point cloud features to achieve high-precision prediction. Experimental results show that with the help of the new representation method, the prediction performance of 3D scene graphs has been significantly improved.
50
Yang, Fan, Mingliang Che, Xinkai Zuo, Lin Li, Jiyi Zhang e Chi Zhang. "Volumetric Representation and Sphere Packing of Indoor Space for Three-Dimensional Room Segmentation". ISPRS International Journal of Geo-Information 10, n. 11 (29 ottobre 2021): 739. http://dx.doi.org/10.3390/ijgi10110739.
Abstract (sommario):
Room segmentation is a basic task for the semantic enrichment of point clouds. Recent studies have mainly projected single-floor point clouds to binary images to realize two-dimensional room segmentation. However, these methods have difficulty solving semantic segmentation problems in complex 3D indoor environments, including cross-floor spaces and rooms inside rooms; this is the bottleneck of indoor 3D modeling for non-Manhattan worlds. To make full use of the abundant geometric and spatial structure information in 3D space, a novel 3D room segmentation method that realizes room segmentation directly in 3D space is proposed in this study. The method utilizes volumetric representation based on a VDB data structure and packs an indoor space with a set of compact spheres to form rooms as separated connected components. Experimental results on different types of indoor point cloud datasets demonstrate the efficiency of the proposed method.