Semantic Point Cloud-Based Adaptive Multiple Object Detection and Tracking for Autonomous Vehicles

LiDAR-based Multiple Object Detection and Tracking (MODT) is one of the essential tasks in autonomous driving. Since MODT is directly related to the safety of an autonomous vehicle, it is critical to provide reliable information about the surrounding objects. For that reason, we propose a semantic p...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Soyeong Kim, Jinsu Ha, Kichun Jo
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Semantic point cloud point cloud semantic segmentation multiple object detection and tracking (MODT) class-adaptive tracking autonomous vehicle LiDAR Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/a2b6e3daa5454c0ba9d4d25f32e31a1b
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:a2b6e3daa5454c0ba9d4d25f32e31a1b
record_format	dspace
spelling	oai:doaj.org-article:a2b6e3daa5454c0ba9d4d25f32e31a1b2021-12-03T00:00:56ZSemantic Point Cloud-Based Adaptive Multiple Object Detection and Tracking for Autonomous Vehicles2169-353610.1109/ACCESS.2021.3130257https://doaj.org/article/a2b6e3daa5454c0ba9d4d25f32e31a1b2021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9627603/https://doaj.org/toc/2169-3536LiDAR-based Multiple Object Detection and Tracking (MODT) is one of the essential tasks in autonomous driving. Since MODT is directly related to the safety of an autonomous vehicle, it is critical to provide reliable information about the surrounding objects. For that reason, we propose a semantic point cloud-based adaptive MODT system for autonomous driving. Semantic point clouds emerge with advances in deep learning-based Point Cloud Semantic Segmentation (PCSS), which assigns semantic information to each point in the point cloud of LiDAR. This semantic information provides several advantages to the MODT system. First, any point corresponding to any static object can be filtered. Because the class information assigned to each point can be directly utilized, filtering is possible without any modeling. Second, the class information of an object can be inferred without any special classification process because the class information is provided from the semantic point cloud. Finally, the clustering and tracking module can consider unique dimensional and dynamic characteristics based on class information. We utilize the Carla simulator and KITTI dataset to verify our method by comparing several existing algorithms. In conclusion, the accuracy of the proposed algorithm is improved by about 176% on average compared to the existing algorithm.Soyeong KimJinsu HaKichun JoIEEEarticleSemantic point cloudpoint cloud semantic segmentationmultiple object detection and tracking (MODT)class-adaptive trackingautonomous vehicleLiDARElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 157550-157562 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Semantic point cloud point cloud semantic segmentation multiple object detection and tracking (MODT) class-adaptive tracking autonomous vehicle LiDAR Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Semantic point cloud point cloud semantic segmentation multiple object detection and tracking (MODT) class-adaptive tracking autonomous vehicle LiDAR Electrical engineering. Electronics. Nuclear engineering TK1-9971 Soyeong Kim Jinsu Ha Kichun Jo Semantic Point Cloud-Based Adaptive Multiple Object Detection and Tracking for Autonomous Vehicles
description	LiDAR-based Multiple Object Detection and Tracking (MODT) is one of the essential tasks in autonomous driving. Since MODT is directly related to the safety of an autonomous vehicle, it is critical to provide reliable information about the surrounding objects. For that reason, we propose a semantic point cloud-based adaptive MODT system for autonomous driving. Semantic point clouds emerge with advances in deep learning-based Point Cloud Semantic Segmentation (PCSS), which assigns semantic information to each point in the point cloud of LiDAR. This semantic information provides several advantages to the MODT system. First, any point corresponding to any static object can be filtered. Because the class information assigned to each point can be directly utilized, filtering is possible without any modeling. Second, the class information of an object can be inferred without any special classification process because the class information is provided from the semantic point cloud. Finally, the clustering and tracking module can consider unique dimensional and dynamic characteristics based on class information. We utilize the Carla simulator and KITTI dataset to verify our method by comparing several existing algorithms. In conclusion, the accuracy of the proposed algorithm is improved by about 176% on average compared to the existing algorithm.
format	article
author	Soyeong Kim Jinsu Ha Kichun Jo
author_facet	Soyeong Kim Jinsu Ha Kichun Jo
author_sort	Soyeong Kim
title	Semantic Point Cloud-Based Adaptive Multiple Object Detection and Tracking for Autonomous Vehicles
title_short	Semantic Point Cloud-Based Adaptive Multiple Object Detection and Tracking for Autonomous Vehicles
title_full	Semantic Point Cloud-Based Adaptive Multiple Object Detection and Tracking for Autonomous Vehicles
title_fullStr	Semantic Point Cloud-Based Adaptive Multiple Object Detection and Tracking for Autonomous Vehicles
title_full_unstemmed	Semantic Point Cloud-Based Adaptive Multiple Object Detection and Tracking for Autonomous Vehicles
title_sort	semantic point cloud-based adaptive multiple object detection and tracking for autonomous vehicles
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/a2b6e3daa5454c0ba9d4d25f32e31a1b
work_keys_str_mv	AT soyeongkim semanticpointcloudbasedadaptivemultipleobjectdetectionandtrackingforautonomousvehicles AT jinsuha semanticpointcloudbasedadaptivemultipleobjectdetectionandtrackingforautonomousvehicles AT kichunjo semanticpointcloudbasedadaptivemultipleobjectdetectionandtrackingforautonomousvehicles
_version_	1718374005831892992

Semantic Point Cloud-Based Adaptive Multiple Object Detection and Tracking for Autonomous Vehicles

Ejemplares similares