Clustering-Based Plane Segmentation Neural Network for Urban Scene Modeling
Urban scene modeling is a challenging but essential task for various applications, such as 3D map generation, city digitization, and AR/VR/metaverse applications. To model man-made structures, such as roads and buildings, which are the major components in general urban scenes, we present a clusterin...
Main Authors: | , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-12-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/21/24/8382 |
_version_ | 1797500761627688960 |
---|---|
author | Hongjae Lee Jiyoung Jung |
author_facet | Hongjae Lee Jiyoung Jung |
author_sort | Hongjae Lee |
collection | DOAJ |
description | Urban scene modeling is a challenging but essential task for various applications, such as 3D map generation, city digitization, and AR/VR/metaverse applications. To model man-made structures, such as roads and buildings, which are the major components in general urban scenes, we present a clustering-based plane segmentation neural network using 3D point clouds, called hybrid K-means plane segmentation (HKPS). The proposed method segments unorganized 3D point clouds into planes by training the neural network to estimate the appropriate number of planes in the point cloud based on hybrid K-means clustering. We consider both the Euclidean distance and cosine distance to cluster nearby points in the same direction for better plane segmentation results. Our network does not require any labeled information for training. We evaluated the proposed method using the Virtual KITTI dataset and showed that our method outperforms conventional methods in plane segmentation. Our code is publicly available. |
first_indexed | 2024-03-10T03:08:29Z |
format | Article |
id | doaj.art-3caf0348c50b4b15a035cc76b38fc989 |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-10T03:08:29Z |
publishDate | 2021-12-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-3caf0348c50b4b15a035cc76b38fc9892023-11-23T10:30:38ZengMDPI AGSensors1424-82202021-12-012124838210.3390/s21248382Clustering-Based Plane Segmentation Neural Network for Urban Scene ModelingHongjae Lee0Jiyoung Jung1Department of Electronic Engineering, Kyung Hee University, Yongin-si 17104, KoreaDepartment of Artificial Intelligence, University of Seoul, Seoul 02504, KoreaUrban scene modeling is a challenging but essential task for various applications, such as 3D map generation, city digitization, and AR/VR/metaverse applications. To model man-made structures, such as roads and buildings, which are the major components in general urban scenes, we present a clustering-based plane segmentation neural network using 3D point clouds, called hybrid K-means plane segmentation (HKPS). The proposed method segments unorganized 3D point clouds into planes by training the neural network to estimate the appropriate number of planes in the point cloud based on hybrid K-means clustering. We consider both the Euclidean distance and cosine distance to cluster nearby points in the same direction for better plane segmentation results. Our network does not require any labeled information for training. We evaluated the proposed method using the Virtual KITTI dataset and showed that our method outperforms conventional methods in plane segmentation. Our code is publicly available.https://www.mdpi.com/1424-8220/21/24/8382point cloud plane extraction3D point clustering3D segmentationurban mapping |
spellingShingle | Hongjae Lee Jiyoung Jung Clustering-Based Plane Segmentation Neural Network for Urban Scene Modeling Sensors point cloud plane extraction 3D point clustering 3D segmentation urban mapping |
title | Clustering-Based Plane Segmentation Neural Network for Urban Scene Modeling |
title_full | Clustering-Based Plane Segmentation Neural Network for Urban Scene Modeling |
title_fullStr | Clustering-Based Plane Segmentation Neural Network for Urban Scene Modeling |
title_full_unstemmed | Clustering-Based Plane Segmentation Neural Network for Urban Scene Modeling |
title_short | Clustering-Based Plane Segmentation Neural Network for Urban Scene Modeling |
title_sort | clustering based plane segmentation neural network for urban scene modeling |
topic | point cloud plane extraction 3D point clustering 3D segmentation urban mapping |
url | https://www.mdpi.com/1424-8220/21/24/8382 |
work_keys_str_mv | AT hongjaelee clusteringbasedplanesegmentationneuralnetworkforurbanscenemodeling AT jiyoungjung clusteringbasedplanesegmentationneuralnetworkforurbanscenemodeling |