Downsampling of a 3D LiDAR Point Cloud by a Tensor Voting Based Method

Ervan, Osman; Temeltaş, Hakan

doi:10.23919/eleco47770.2019.8990544

Cited by 4 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Given our results, previous methods cannot handle noisy, large-scale and density-varying point clouds when they are used to reduce the number of points. Eventually, similar to our method but subsequent to our previous work (Labussière et al, 2018), Ervan and Temeltas (2019) also proposed a sampling algorithm based on a modified tensor voting framework to preserve geometric primitives while down sampling dense areas taking most salient points as representatives. Only qualitative results are given on only one scan, and no comparison have been made.…”

Section: Related Workmentioning

confidence: 81%

Geometry Preserving Sampling Method Based on Spectral Decomposition for Large-Scale Environments

2020

View full text Add to dashboard Cite

In the context of 3D mapping, larger and larger point clouds are acquired with lidar sensors. Although pleasing to the eye, dense maps are not necessarily tailored for practical applications. For instance, in a surface inspection scenario, keeping geometric information such as the edges of objects is essential to detect cracks, whereas very dense areas of very little information such as the ground could hinder the main goal of the application. Several strategies exist to address this problem by reducing the number of points. However, they tend to underperform with non-uniform density, large sensor noise, spurious measurements, and large-scale point clouds, which is the case in mobile robotics. This paper presents a novel sampling algorithm based on spectral decomposition analysis to derive local density measures for each geometric primitive. The proposed method, called Spectral Decomposition Filter (SpDF), identifies and preserves geometric information along the topology of point clouds and is able to scale to large environments with a non-uniform density. Finally, qualitative and quantitative experiments verify the feasibility of our method and present a large-scale evaluation of SpDF with other seven point cloud sampling algorithms, in the context of the 3D registration problem using the Iterative Closest Point (ICP) algorithm on real-world datasets. Results show that a compression ratio up to 97 % can be achieved when accepting a registration error within the range accuracy of the sensor, here for large scale environments of less than 2 cm.

show abstract

Section: Related Workmentioning

confidence: 81%

Geometry Preserving Sampling Method Based on Spectral Decomposition for Large-Scale Environments

2020

View full text Add to dashboard Cite

show abstract

“…However, considering its successful results, it has been seen as a suitable method for benchmarking. Ervan and Temeltas (2019) proposed a sampling method by using the tensor voting concept to find out the dense regions of the point cloud and apply the sampling algorithm to those parts to preserve the geometry. Ervan and Temeltas (2020) extended the tensor voting‐based sampling method into point cloud registration, with the main focus remaining on the preservation of geometric features.…”

Section: Related Workmentioning

confidence: 99%

A histogram‐based sampling method for point cloud registration

Ervan

Temeltaş

2023

The Photogrammetric Record

Self Cite

View full text Add to dashboard Cite

Accurate and efficient point cloud registration is essential in various fields, such as robotics, autonomous driving and medical imaging. The size of point clouds presents a significant challenge for existing registration methods. In this paper, a novel point cloud sampling method to improve the performance of the point cloud registration process is proposed. Instead of geometric feature preservation, which is preferred in most existing sampling methods, our approach scales every point and groups the scaled points into clusters to generate a histogram for the point cloud. The histogram is then used to identify the most significant regions of the point cloud to create the downsampled output data. Experimental results demonstrate that the proposed method improves accuracy and is robust against noise. Registration errors are reduced by up to 7% in rotation and 116% in translation. Additionally, the proposed method filtered out up to 98% of noise from the point cloud that was uniformly distributed at a rate of 25%.

show abstract

“…The surface reconstruction algorithm (Labatut et al, 2009) mainly includes Poisson or Delauny network construction, and mesh texture mapping (Allene et al, 2008) assigns two-dimensional (2D) space point information (such as colour and brightness) to the 3D space points in the object space through a certain mapping relationship (Ervan & Temeltas, 2023). The photometric consistency of the model is maintained by requiring uniform light between blocks (Luo, 2015).…”

Section: Rel Ated Workmentioning

confidence: 99%

Rapid 3D modelling: Clustering method based on dynamic load balancing strategy

Ge,

Guo,

et al. 2023

The Photogrammetric Record

View full text Add to dashboard Cite

Three‐dimensional (3D) reconstruction is a pivotal research area within computer vision and photogrammetry, offering a valuable foundation of data for the development of smart cities. However, existing methods for constructing 3D models from unmanned aerial vehicle (UAV) images often suffer from slow processing speeds and low central processing unit (CPU)/graphics processing unit (GPU) utilization rates. Furthermore, the utilization of cluster‐based distributed computing for 3D modelling frequently results in inefficient resource allocation across nodes. To address these challenges, this paper presents a novel approach to 3D modelling in clusters, incorporating a dynamic load‐balancing strategy. The method divides the 3D reconstruction process into multiple stages to lay the groundwork for distributing tasks across multiple nodes efficiently. Instead of traditional traversal‐based communication, this approach employs gossip communication techniques to reduce the network overhead. To boost the modelling efficiency, a dynamic load‐balancing strategy is introduced that prevents nodes from becoming overloaded, thus optimizing resource usage during the modelling process and alleviating resource waste issues in multidevice clusters. The experimental results indicate that in small‐scale data environments, this approach improves CPU/GPU utilization by 35.8%/23.4% compared with single‐machine utilization. In large‐scale data environments for cluster‐based 3D modelling tests, this method enhances the average efficiency by 61.4% compared with traditional 3D modelling software while maintaining a comparable model accuracy. In computer vision and photogrammetry, research enhances 3D reconstruction for smart cities. To address slow UAV‐based methods, the study employs dynamic load balancing and ‘gossip’ communication to minimize network overhead. In small data tests, the approach improves CPU and GPU efficiency by 20.7% and 40.3%, respectively. In large data settings, it outperforms existing methods by 61.38% while maintaining accuracy.

show abstract

Downsampling of a 3D LiDAR Point Cloud by a Tensor Voting Based Method

Cited by 4 publications

References 13 publications

Geometry Preserving Sampling Method Based on Spectral Decomposition for Large-Scale Environments

Geometry Preserving Sampling Method Based on Spectral Decomposition for Large-Scale Environments

A histogram‐based sampling method for point cloud registration

Rapid 3D modelling: Clustering method based on dynamic load balancing strategy

Contact Info

Product

Resources

About