The Impact of the Number of k-Means Clusters on 3D Point Cloud Registration

Peter Ankomah; Peter Vangorp

doi:10.24132/CSRN.3401.2

The Impact of the Number of k-Means Clusters on 3D Point Cloud Registration

Research output: Contribution to conference › Paper › Academic

Abstract

Point cloud registration plays a crucial role in many applications, from robotics and autonomous navigation to
medical imaging and 3D scene reconstruction. While the Iterative Closest Point (ICP) algorithm is a well-known
shape registration choice, its efficiency and accuracy can be affected by the vast search space for point correspondences. k-means clustering emerges as a promising solution for partitioning the search space into smaller clusters
to reduce the computational complexity and increase the performance of the matching. However, the number and
size of these clusters and how they affect the registration remains a critical and yet not fully explored factor. This
paper delves into the relationship between the number of k-means clusters and point cloud registration accuracy.
To determine the effect of the number of k-means clusters on registration accuracy and efficiency and to understand
any emerging pattern, k-meansICP is developed to use the k-means algorithm to cluster the correspondence search
space. Two sets of 3D molecular shapes with differing complexities are matched using initial rotation angles 15,
30, and 60 degrees with 2 to 10 k-means clusters. The results are then compared with the original ICP algorithm.

Original language	English
Pages	3-12
Number of pages	10
DOIs	https://doi.org/10.24132/CSRN.3401.2
Publication status	Published - 4 Jun 2024
Event	International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision 2024 - Primavera Hotel & Congress Center, Pilsen, Czech Republic Duration: 3 Jun 2024 → 6 Jun 2024 Conference number: 32

Conference

Conference	International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision 2024
Abbreviated title	WSCG 2024
Country/Territory	Czech Republic
City	Pilsen
Period	3/06/24 → 6/06/24

Keywords

3D Point Cloud
3D Shape Registration
Iterative Closest Point
k-Means Clustering

Access to Document

10.24132/CSRN.3401.2

2024_A05-full
Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.
Final published version, 1.01 MB

Cite this

@conference{619c1835531143a08f1236ece183ca77,

title = "The Impact of the Number of k-Means Clusters on 3D Point Cloud Registration",

abstract = "Point cloud registration plays a crucial role in many applications, from robotics and autonomous navigation tomedical imaging and 3D scene reconstruction. While the Iterative Closest Point (ICP) algorithm is a well-knownshape registration choice, its efficiency and accuracy can be affected by the vast search space for point correspondences. k-means clustering emerges as a promising solution for partitioning the search space into smaller clustersto reduce the computational complexity and increase the performance of the matching. However, the number andsize of these clusters and how they affect the registration remains a critical and yet not fully explored factor. Thispaper delves into the relationship between the number of k-means clusters and point cloud registration accuracy.To determine the effect of the number of k-means clusters on registration accuracy and efficiency and to understandany emerging pattern, k-meansICP is developed to use the k-means algorithm to cluster the correspondence searchspace. Two sets of 3D molecular shapes with differing complexities are matched using initial rotation angles 15,30, and 60 degrees with 2 to 10 k-means clusters. The results are then compared with the original ICP algorithm.",

keywords = "3D Point Cloud, 3D Shape Registration, Iterative Closest Point, k-Means Clustering",

author = "Peter Ankomah and Peter Vangorp",

note = "Publisher Copyright: {\textcopyright} 2024 University of West Bohemia. All rights reserved.; International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision 2024, WSCG 2024 ; Conference date: 03-06-2024 Through 06-06-2024",

year = "2024",

month = jun,

day = "4",

doi = "10.24132/CSRN.3401.2",

language = "English",

pages = "3--12",

}

TY - CONF

T1 - The Impact of the Number of k-Means Clusters on 3D Point Cloud Registration

AU - Ankomah, Peter

AU - Vangorp, Peter

N1 - Conference code: 32

PY - 2024/6/4

Y1 - 2024/6/4

N2 - Point cloud registration plays a crucial role in many applications, from robotics and autonomous navigation tomedical imaging and 3D scene reconstruction. While the Iterative Closest Point (ICP) algorithm is a well-knownshape registration choice, its efficiency and accuracy can be affected by the vast search space for point correspondences. k-means clustering emerges as a promising solution for partitioning the search space into smaller clustersto reduce the computational complexity and increase the performance of the matching. However, the number andsize of these clusters and how they affect the registration remains a critical and yet not fully explored factor. Thispaper delves into the relationship between the number of k-means clusters and point cloud registration accuracy.To determine the effect of the number of k-means clusters on registration accuracy and efficiency and to understandany emerging pattern, k-meansICP is developed to use the k-means algorithm to cluster the correspondence searchspace. Two sets of 3D molecular shapes with differing complexities are matched using initial rotation angles 15,30, and 60 degrees with 2 to 10 k-means clusters. The results are then compared with the original ICP algorithm.

AB - Point cloud registration plays a crucial role in many applications, from robotics and autonomous navigation tomedical imaging and 3D scene reconstruction. While the Iterative Closest Point (ICP) algorithm is a well-knownshape registration choice, its efficiency and accuracy can be affected by the vast search space for point correspondences. k-means clustering emerges as a promising solution for partitioning the search space into smaller clustersto reduce the computational complexity and increase the performance of the matching. However, the number andsize of these clusters and how they affect the registration remains a critical and yet not fully explored factor. Thispaper delves into the relationship between the number of k-means clusters and point cloud registration accuracy.To determine the effect of the number of k-means clusters on registration accuracy and efficiency and to understandany emerging pattern, k-meansICP is developed to use the k-means algorithm to cluster the correspondence searchspace. Two sets of 3D molecular shapes with differing complexities are matched using initial rotation angles 15,30, and 60 degrees with 2 to 10 k-means clusters. The results are then compared with the original ICP algorithm.

KW - 3D Point Cloud

KW - 3D Shape Registration

KW - Iterative Closest Point

KW - k-Means Clustering

U2 - 10.24132/CSRN.3401.2

DO - 10.24132/CSRN.3401.2

M3 - Paper

SP - 3

EP - 12

T2 - International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision 2024

Y2 - 3 June 2024 through 6 June 2024

ER -

The Impact of the Number of k-Means Clusters on 3D Point Cloud Registration

Abstract

Conference

Keywords

Access to Document

Fingerprint

Cite this