Convolutional neural network for 3D point clouds matching

2021 ◽  
Author(s):  
Sergei Voronin ◽  
Artyom Makovetskii ◽  
Aleksei Voronin ◽  
Dmitrii Zhernov
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Point Clouds ◽  
3D Point Clouds

scholarly journals Convolutional Neural Network for Extracting 3D Point Clouds of Fibrous Web From Multi-Focus Images

IEEE Access ◽  
2020 ◽  
Vol 8 ◽  
pp. 87857-87869
Author(s):  
Jue Hou ◽  
Wenbin Ouyang ◽  
Bugao Xu ◽  
Rongwu Wang
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Point Clouds ◽  
3D Point Clouds

scholarly journals 3D Point Cloud Recognition Based on a Multi-View Convolutional Neural Network

Sensors ◽  
2018 ◽  
Vol 18 (11) ◽  
pp. 3681 ◽  
Author(s):  
Le Zhang ◽  
Jian Sun ◽  
Qiang Zheng
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Point Cloud ◽  
Recognition Performance ◽  
Three Dimensional ◽  
Point Clouds ◽  
Feature Descriptor ◽  
Cloud Data ◽  
Single View ◽  
3D Point Clouds

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.

A Dense RGB-D SLAM Algorithm based on Convolutional Neural Network of Multi-layer Image Invariant Feature

2021 ◽  
Author(s):  
SU YAN ◽  
Lei Yu
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Pose Estimation ◽  
Autonomous Vehicles ◽  
Reconstruction Algorithm ◽  
Point Clouds ◽  
Feature Point ◽  
3D Point Clouds ◽  
Invariant Feature ◽  
Layer Image

Abstract Simultaneous Localization and Mapping (SLAM) is one of the key technologies used in sweepers, autonomous vehicles, virtual reality and other fields. This paper presents a dense RGB-D SLAM reconstruction algorithm based on convolutional neural network of multi-layer image invariant feature transformation. The main contribution of the system lies in the construction of a convolutional neural network based on multi-layer image invariant feature, which optimized the extraction of ORB (Oriented FAST and Rotated Brief) feature points and the reconstruction effect. After the feature point matching, pose estimation, loop detection and other steps, the 3D point clouds were finally spliced to construct a complete and smooth spatial model. The system can improve the accuracy and robustness in feature point processing and pose estimation. Comparative experiments show that the optimized algorithm saves 0.093s compared to the ordinary extraction algorithm while guaranteeing a high accuracy rate at the same time. The results of reconstruction experiments show that the spatial models have more clear details, smoother connection with no fault layers than the original ones. The reconstruction results are generally better than other common algorithms, such as Kintinuous, Elasticfusion and ORBSLAM2 dense reconstruction.

Real-Time 3D object detection using improved convolutional neural network based on image-driven point cloud

2021 ◽  
Vol 14 ◽  
Author(s):  
Zhiyong Gao ◽  
Jianhong Xiang
Keyword(s):  
Neural Network ◽  
Object Detection ◽  
Convolutional Neural Network ◽  
Real Time ◽  
Point Cloud ◽  
Point Clouds ◽  
3D Point Cloud ◽  
3D Object ◽  
3D Object Detection ◽  
Instance Segmentation

Background: While detecting the object directly from the 3D point cloud, the natural 3D patterns and invariance of 3D data are often obscure. Objective: In this work, we aimed at studying the 3D object detection from discrete, disordered and sparse 3D point clouds. Methods: The CNN is composed of the frustum sequence module, 3D instance segmentation module S-NET, 3D point cloud transformation module T-NET, and 3D boundary box estimation module E-NET. The search space of the object is determined by the frustum sequence module. The instance segmentation of the point cloud is performed by the 3D instance segmentation module. The 3D coordinates of the object are confirmed by the transformation module and the 3D bounding box estimation module. Results: Evaluated on KITTI benchmark dataset, our method outperforms the state of the art by remarkable margins while having real-time capability. Conclusion: We achieve real-time 3D object detection by proposing an improved convolutional neural network (CNN) based on image-driven point clouds.

scholarly journals Segmentation and Multi-Scale Convolutional Neural Network-Based Classification of Airborne Laser Scanner Data

Sensors ◽  
2018 ◽  
Vol 18 (10) ◽  
pp. 3347 ◽  
Author(s):  
Zhishuang Yang ◽  
Bo Tan ◽  
Huikun Pei ◽  
Wanshou Jiang
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Laser Scanning ◽  
Laser Scanner ◽  
Region Growing ◽  
Point Clouds ◽  
Classification Method ◽  
Multi Scale ◽  
Airborne Laser

The classification of point clouds is a basic task in airborne laser scanning (ALS) point cloud processing. It is quite a challenge when facing complex observed scenes and irregular point distributions. In order to reduce the computational burden of the point-based classification method and improve the classification accuracy, we present a segmentation and multi-scale convolutional neural network-based classification method. Firstly, a three-step region-growing segmentation method was proposed to reduce both under-segmentation and over-segmentation. Then, a feature image generation method was used to transform the 3D neighborhood features of a point into a 2D image. Finally, feature images were treated as the input of a multi-scale convolutional neural network for training and testing tasks. In order to obtain performance comparisons with existing approaches, we evaluated our framework using the International Society for Photogrammetry and Remote Sensing Working Groups II/4 (ISPRS WG II/4) 3D labeling benchmark tests. The experiment result, which achieved 84.9% overall accuracy and 69.2% of average F1 scores, has a satisfactory performance over all participating approaches analyzed.

scholarly journals Extraction of Guardrails from MMS Data Using Convolutional Neural Network

2021 ◽  
Vol 15 (3) ◽  
pp. 258-267
Author(s):  
Hiroki Matsumoto ◽  
◽  
Yuma Mori ◽  
Hiroshi Masuda
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Asset Management ◽  
Extraction Method ◽  
Point Clouds ◽  
Mobile Mapping ◽  
Method Point ◽  
Mobile Mapping Systems ◽  
Capture Point ◽  
Mapping Systems

Mobile mapping systems can capture point clouds and digital images of roadside objects. Such data are useful for maintenance, asset management, and 3D map creation. In this paper, we discuss methods for extracting guardrails that separate roadways and walkways. Since there are various shape patterns for guardrails in Japan, flexible methods are required for extracting them. We propose a new extraction method based on point processing and a convolutional neural network (CNN). In our method, point clouds and images are segmented into small fragments, and their features are extracted using CNNs for images and point clouds. Then, features from images and point clouds are combined and investigated using whether they are guardrails or not. Based on our experiments, our method could extract guardrails from point clouds with a high success rate.

A 3-D Lightweight Convolutional Neural Network for Detecting Docking Structures in Cluttered Environments

2021 ◽  
Vol 55 (4) ◽  
pp. 88-98
Author(s):  
Maria Inês Pereira ◽  
Pedro Nuno Leite ◽  
Andry Maykol Pinto
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Research Work ◽  
Point Clouds ◽  
Measurement Unit ◽  
Maritime Industry ◽  
Worst Case ◽  
Cluttered Environments ◽  
Autonomous Surface Vehicles ◽  
Average Accuracy

Abstract The maritime industry has been following the paradigm shift toward the automation of typically intelligent procedures, with research regarding autonomous surface vehicles (ASVs) having seen an upward trend in recent years. However, this type of vehicle cannot be employed on a full scale until a few challenges are solved. For example, the docking process of an ASV is still a demanding task that currently requires human intervention. This research work proposes a volumetric convolutional neural network (vCNN) for the detection of docking structures from 3-D data, developed according to a balance between precision and speed. Another contribution of this article is a set of synthetically generated data regarding the context of docking structures. The dataset is composed of LiDAR point clouds, stereo images, GPS, and Inertial Measurement Unit (IMU) information. Several robustness tests carried out with different levels of Gaussian noise demonstrated an average accuracy of 93.34% and a deviation of 5.46% for the worst case. Furthermore, the system was fine-tuned and evaluated in a real commercial harbor, achieving an accuracy of over 96%. The developed classifier is able to detect different types of structures and works faster than other state-of-the-art methods that establish their performance in real environments.

Sign in / Sign up

Export Citation Format

Share Document