Dynamic Obstacles Rejection for 3D Map Simultaneous Updating

Shi, Wenjun; Li, Jiamao; Liu, Yanqing; Zhu, Dongchen; Yang, Dongdong; Zhang, Xiaolin

doi:10.1109/access.2018.2836192

Cited by 13 publications

(5 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Of the direct simplification methods, spatial index is an important method to process point cloud. Shi et al [12] used kd-tree to obtain spurious trails and updated point cloud map. Goswami et al [13] used kd-tree to process points.…”

Section: Introductionmentioning

confidence: 99%

A Novel Simplification Method for 3D Geometric Point Cloud Based on the Importance of Point

Liu

Fan

et al. 2019

IEEE Access

View full text Add to dashboard Cite

3D point cloud simplification is an important pretreatment in surface reconstruction for sparing computer resources and improving reconstruction speed. However, existing methods often sacrifice the simplification precision to improve the simplification speed, or sacrifice the speed to improve precision. A proper balance between the simplification speed and the simplification accuracy is still a challenge. In this paper, we propose a new simplification method based on the importance of point. Named as detail feature points simplified algorithm (DFPSA), this algorithm has distinct processes to achieve improvements in three aspects. First, a rule of k neighborhood search is set to ensure the points found are the closest to the sample point. In this way, the accuracy of calculated normal vector of the point cloud is significantly improved, and the search speed is largely increased. Second, a formula that considers multiple characteristics for measuring the importance of point is proposed. Thereupon, the main detail features of the point cloud are preserved. Finally, an octree structure is employed to simplify the remaining points, through which holes in reconstructing point cloud are obviously reduced. The DFPSA is applied to four different data sets, and the corresponding results are compared with those of other five algorithms. The experimental results demonstrate that the DFPSA brings better simplification effects than existing counterparts, and the DFPSA not only can simplify point cloud but also has good effect in simplifying subject's narrow contours.INDEX TERMS 3D geometric point cloud simplification, detail feature point, k neighborhood search rule, the importance of point.

show abstract

Section: Introductionmentioning

confidence: 99%

A Novel Simplification Method for 3D Geometric Point Cloud Based on the Importance of Point

Liu

Fan

et al. 2019

IEEE Access

View full text Add to dashboard Cite

show abstract

“…The following studies are related to real time dynamic object location recognition. Wenjun (2018) created an algorithm that updates the map by eliminating dynamic peripheral trails in the point cloud through object tracking [8]. By distinguishing between a static background and moving object, only the static environment was reconstructed in three dimensions which could be applied to the simultaneous localization and mapping (SLAM) system.…”

Section: Related Workmentioning

confidence: 99%

Road Dynamic Object Mapping System Based on Edge-Fog-Cloud Computing

2021

View full text Add to dashboard Cite

Dynamic objects appearing on the road without notice can cause serious accidents. However, the detection ranges of roadside unit and CCTV that collect current road information are very limited. Moreover, there are a lack of systems for managing the collected information. In this study, a dynamic mapping system was implemented using a connected car that collected road environments data continuously. Additionally, edge-fog-cloud computing was applied to efficiently process large amounts of road data. For accurate dynamic mapping, the following steps are proposed: first, the classification and 3D position of road objects are estimated through a stereo camera and GPS data processing, and the coordinates of objects are mapped to a preset grid cell. Second, object information is transmitted in real time to a constructed big data processing platform. Subsequently, the collected information is compared with the grid information of an existing map, and the map is updated. As a result, an accurate dynamic map is created and maintained. In addition, this study verifies that maps can be shared in real time with IoT devices in various network environments, and this can support a safe driving milieu.

show abstract

“…Our approach directly compares point clouds among frames and is related to the work of Yoon et al [57] and Shi et al [49]. Yoon et al [57] work with LiDAR data and need to introduce a time-lag of one frame to reliably cope with occlusions.…”

Section: Related Workmentioning

confidence: 99%

“…We expand their idea to handle noisy stereo camera data with potentially incomplete depth information and additionally introduce a novel approach to differentiate between dynamic and previously occluded points without the need of a time-lag of one frame. Shi et al [49] use RGB-D data and remove points during dense reconstruction in case they are spatially inconsistent between frames. In contrast, our method is able to classify all points of an object as dynamic, even though only their subset shows spatial inconsistency.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Leveraging Stereo-Camera Data for Real-Time Dynamic Obstacle Detection and Tracking

Eppenberger¹,

Cesari²,

Dymczyk³

et al. 2020

Preprint

View full text Add to dashboard Cite

Dynamic obstacle avoidance is one crucial component for compliant navigation in crowded environments. In this paper we present a system for accurate and reliable detection and tracking of dynamic objects using noisy point cloud data generated by stereo cameras. Our solution is realtime capable and specifically designed for the deployment on computationally-constrained unmanned ground vehicles. The proposed approach identifies individual objects in the robot's surroundings and classifies them as either static or dynamic. The dynamic objects are labeled as either a person or a generic dynamic object. We then estimate their velocities to generate a 2D occupancy grid that is suitable for performing obstacle avoidance. We evaluate the system in indoor and outdoor scenarios and achieve real-time performance on a consumergrade computer. On our test-dataset, we reach a MOTP of 0.07 ± 0.07m, and a MOTA of 85.3% for the detection and tracking of dynamic objects. We reach a precision of 96.9% for the detection of static objects.

show abstract

Dynamic Obstacles Rejection for 3D Map Simultaneous Updating

Cited by 13 publications

References 20 publications

A Novel Simplification Method for 3D Geometric Point Cloud Based on the Importance of Point

A Novel Simplification Method for 3D Geometric Point Cloud Based on the Importance of Point

Road Dynamic Object Mapping System Based on Edge-Fog-Cloud Computing

Leveraging Stereo-Camera Data for Real-Time Dynamic Obstacle Detection and Tracking

Contact Info

Product

Resources

About