scholarly journals ERASOR: Egocentric Ratio of Pseudo Occupancy-Based Dynamic Object Removal for Static 3D Point Cloud Map Building

2021 ◽  
Vol 6 (2) ◽  
pp. 2272-2279
Author(s):  
Hyungtae Lim ◽  
Sungwon Hwang ◽  
Hyun Myung
Keyword(s):  
Point Cloud ◽  
Map Building ◽  
3D Point Cloud ◽  
Dynamic Object ◽  
Object Removal

3D LiDAR Point Cloud Loop Detection Based on Dynamic Object Removal

2021 ◽  
Author(s):  
Panyun Ding ◽  
Zhihui Wang
Keyword(s):  
Point Cloud ◽  
Dynamic Object ◽  
Object Removal ◽  
Loop Detection ◽  
3D Lidar

scholarly journals A FAST AND FLEXIBLE METHOD FOR META-MAP BUILDING FOR ICP BASED SLAM

2016 ◽  
Vol XLI-B3 ◽  
pp. 273-278
Author(s):  
A. Kurian ◽  
K. W. Morin
Keyword(s):  
Real Time ◽  
Point Cloud ◽  
Sampling Strategy ◽  
Cost Effective ◽  
Point Clouds ◽  
Limited Range ◽  
Map Building ◽  
3D Point Cloud ◽  
Circular Buffer ◽  
Localization And Mapping

Recent developments in LiDAR sensors make mobile mapping fast and cost effective. These sensors generate a large amount of data which in turn improves the coverage and details of the map. Due to the limited range of the sensor, one has to collect a series of scans to build the entire map of the environment. If we have good GNSS coverage, building a map is a well addressed problem. But in an indoor environment, we have limited GNSS reception and an inertial solution, if available, can quickly diverge. In such situations, simultaneous localization and mapping (SLAM) is used to generate a navigation solution and map concurrently. SLAM using point clouds possesses a number of computational challenges even with modern hardware due to the shear amount of data. In this paper, we propose two strategies for minimizing the cost of computation and storage when a 3D point cloud is used for navigation and real-time map building. We have used the 3D point cloud generated by Leica Geosystems's Pegasus Backpack which is equipped with Velodyne VLP-16 LiDARs scanners. To improve the speed of the conventional iterative closest point (ICP) algorithm, we propose a point cloud sub-sampling strategy which does not throw away any key features and yet significantly reduces the number of points that needs to be processed and stored. In order to speed up the correspondence finding step, a dual kd-tree and circular buffer architecture is proposed. We have shown that the proposed method can run in real time and has excellent navigation accuracy characteristics.

Robust Method for Static 3D Point Cloud Map Building using Multi-View Images with Multi-Resolution

2021 ◽  
Author(s):  
Chen Yao ◽  
Hu Zhu ◽  
Shipeng Lv ◽  
Dingyuan Zhang ◽  
Zhenzhong Jia
Keyword(s):  
Point Cloud ◽  
Map Building ◽  
3D Point Cloud ◽  
Robust Method

scholarly journals A FAST AND FLEXIBLE METHOD FOR META-MAP BUILDING FOR ICP BASED SLAM

2016 ◽  
Vol XLI-B3 ◽  
pp. 273-278 ◽  
Author(s):  
A. Kurian ◽  
K. W. Morin
Keyword(s):  
Real Time ◽  
Point Cloud ◽  
Sampling Strategy ◽  
Cost Effective ◽  
Point Clouds ◽  
Limited Range ◽  
Map Building ◽  
3D Point Cloud ◽  
Circular Buffer ◽  
Localization And Mapping

Recent developments in LiDAR sensors make mobile mapping fast and cost effective. These sensors generate a large amount of data which in turn improves the coverage and details of the map. Due to the limited range of the sensor, one has to collect a series of scans to build the entire map of the environment. If we have good GNSS coverage, building a map is a well addressed problem. But in an indoor environment, we have limited GNSS reception and an inertial solution, if available, can quickly diverge. In such situations, simultaneous localization and mapping (SLAM) is used to generate a navigation solution and map concurrently. SLAM using point clouds possesses a number of computational challenges even with modern hardware due to the shear amount of data. In this paper, we propose two strategies for minimizing the cost of computation and storage when a 3D point cloud is used for navigation and real-time map building. We have used the 3D point cloud generated by Leica Geosystems's Pegasus Backpack which is equipped with Velodyne VLP-16 LiDARs scanners. To improve the speed of the conventional iterative closest point (ICP) algorithm, we propose a point cloud sub-sampling strategy which does not throw away any key features and yet significantly reduces the number of points that needs to be processed and stored. In order to speed up the correspondence finding step, a dual kd-tree and circular buffer architecture is proposed. We have shown that the proposed method can run in real time and has excellent navigation accuracy characteristics.

Analysis of Efficient Visualization and Interactive Editing of 3D Point Cloud Data

2016 ◽  
Vol 3 (2) ◽  
pp. 32
Author(s):  
Rekha Patil ◽  
Ashok Kumar Patil ◽  
Youngho Chai
Keyword(s):  
Point Cloud ◽  
3D Point Cloud ◽  
Point Cloud Data ◽  
Cloud Data ◽  
Interactive Editing

scholarly journals Label3DMaize: toolkit for 3D point cloud data annotation of maize shoots

GigaScience ◽  
2021 ◽  
Vol 10 (5) ◽  
Author(s):  
Teng Miao ◽  
Weiliang Wen ◽  
Yinglun Li ◽  
Sheng Wu ◽  
Chao Zhu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Point Cloud ◽  
Training Dataset ◽  
Growth Stages ◽  
3D Point Cloud ◽  
Plant Structure ◽  
Data Annotation ◽  
Cloud Data ◽  
Point Cloud Segmentation ◽  
Different Growth Stages

Abstract Background The 3D point cloud is the most direct and effective data form for studying plant structure and morphology. In point cloud studies, the point cloud segmentation of individual plants to organs directly determines the accuracy of organ-level phenotype estimation and the reliability of the 3D plant reconstruction. However, highly accurate, automatic, and robust point cloud segmentation approaches for plants are unavailable. Thus, the high-throughput segmentation of many shoots is challenging. Although deep learning can feasibly solve this issue, software tools for 3D point cloud annotation to construct the training dataset are lacking. Results We propose a top-to-down point cloud segmentation algorithm using optimal transportation distance for maize shoots. We apply our point cloud annotation toolkit for maize shoots, Label3DMaize, to achieve semi-automatic point cloud segmentation and annotation of maize shoots at different growth stages, through a series of operations, including stem segmentation, coarse segmentation, fine segmentation, and sample-based segmentation. The toolkit takes ∼4–10 minutes to segment a maize shoot and consumes 10–20% of the total time if only coarse segmentation is required. Fine segmentation is more detailed than coarse segmentation, especially at the organ connection regions. The accuracy of coarse segmentation can reach 97.2% that of fine segmentation. Conclusion Label3DMaize integrates point cloud segmentation algorithms and manual interactive operations, realizing semi-automatic point cloud segmentation of maize shoots at different growth stages. The toolkit provides a practical data annotation tool for further online segmentation research based on deep learning and is expected to promote automatic point cloud processing of various plants.

scholarly journals Estimating Reference Bony Shape Models for Orthognathic Surgical Planning Using 3D Point-Cloud Deep Learning

2021 ◽  
pp. 1-1
Author(s):  
Deqiang Xiao ◽  
Chunfeng Lian ◽  
Han Deng ◽  
Tianshu Kuang ◽  
Qin Liu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Point Cloud ◽  
Surgical Planning ◽  
3D Point Cloud ◽  
Shape Models

GridNet: efficiently learning deep hierarchical representation for 3D point cloud understanding

2021 ◽  
Vol 16 (1) ◽  
Author(s):  
Huiqun Wang ◽  
Di Huang ◽  
Yunhong Wang
Keyword(s):  
Point Cloud ◽  
3D Point Cloud ◽  
Hierarchical Representation

Adaptive Surface Fitting Convolution for 3D Point Cloud Analysis

2020 ◽  
Author(s):  
Hezhi Cao ◽  
Yanxin Ma ◽  
Ronghui Zhan ◽  
Chao Ma ◽  
Jun Zhang
Keyword(s):  
Point Cloud ◽  
Surface Fitting ◽  
3D Point Cloud ◽  
Point Cloud Analysis ◽  
Cloud Analysis
Sign in / Sign up

Export Citation Format

Share Document