Track-Cued Radar Point Cloud Target Classification

Lihui Chen; Mujtaba Hassan; Satish Ravindran; Ryan Wu

doi:10.1109/IEEECONF60004.2024.10942901

Track-Cued Radar Point Cloud Target Classification

Lihui Chen, Mujtaba Hassan, Satish Ravindran, Ryan Wu

Microwave Sensing, Signals & Systems

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

6 Downloads (Pure)

Abstract

A novel temporal-spatial object classification neural network model is proposed to improve the classification capability of tracked objects. It takes queued points of tracked objects using multiple frames as input, utilizes spatial and temporal information from these points for sampling and grouping as well as extracts hierarchical temporal-spatial features for target classification. Experimental results on a proprietary 4D Imaging Radar dataset and open-source 2D RadarScenes dataset demonstrate that the proposed tracker-cued radar point-cloud target classification method allows the model to learn meaningful appearance and motion features from sparse radar points data, and achieves accurate classification output as compared to a baseline method, while being efficient to run on edge hardware with limited resources.

Original language	English
Title of host publication	Conference Record of the 58th Asilomar Conference on Signals, Systems and Computers, ACSSC 2024
Editors	Michael B. Matthews
Publisher	IEEE
Pages	1354-1359
Number of pages	6
ISBN (Electronic)	9798350354058
DOIs	https://doi.org/10.1109/IEEECONF60004.2024.10942901
Publication status	Published - 2024
Event	58th Asilomar Conference on Signals, Systems and Computers, ACSSC 2024 - Hybrid, Pacific Grove, United States Duration: 27 Oct 2024 → 30 Oct 2024

Publication series

Name	Conference Record - Asilomar Conference on Signals, Systems and Computers
ISSN (Print)	1058-6393

Conference

Conference	58th Asilomar Conference on Signals, Systems and Computers, ACSSC 2024
Country/Territory	United States
City	Hybrid, Pacific Grove
Period	27/10/24 → 30/10/24

Bibliographical note

Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

Object classification
temporal-spatial feature extraction
temporal-spatial grouping
temporal-spatial sampling

Access to Document

10.1109/IEEECONF60004.2024.10942901

Track-Cued_Radar_Point_Cloud_Target_ClassificationFinal published version, 0.99 MB

Cite this

Chen, L., Hassan, M., Ravindran, S., & Wu, R. (2024). Track-Cued Radar Point Cloud Target Classification. In M. B. Matthews (Ed.), Conference Record of the 58th Asilomar Conference on Signals, Systems and Computers, ACSSC 2024 (pp. 1354-1359). (Conference Record - Asilomar Conference on Signals, Systems and Computers). IEEE. https://doi.org/10.1109/IEEECONF60004.2024.10942901

@inproceedings{941303b9ee0340138a888a012fc11551,

title = "Track-Cued Radar Point Cloud Target Classification",

abstract = "A novel temporal-spatial object classification neural network model is proposed to improve the classification capability of tracked objects. It takes queued points of tracked objects using multiple frames as input, utilizes spatial and temporal information from these points for sampling and grouping as well as extracts hierarchical temporal-spatial features for target classification. Experimental results on a proprietary 4D Imaging Radar dataset and open-source 2D RadarScenes dataset demonstrate that the proposed tracker-cued radar point-cloud target classification method allows the model to learn meaningful appearance and motion features from sparse radar points data, and achieves accurate classification output as compared to a baseline method, while being efficient to run on edge hardware with limited resources.",

keywords = "Object classification, temporal-spatial feature extraction, temporal-spatial grouping, temporal-spatial sampling",

author = "Lihui Chen and Mujtaba Hassan and Satish Ravindran and Ryan Wu",

note = "Green Open Access added to TU Delft Institutional Repository {\textquoteleft}You share, we take care!{\textquoteright} – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.; 58th Asilomar Conference on Signals, Systems and Computers, ACSSC 2024 ; Conference date: 27-10-2024 Through 30-10-2024",

year = "2024",

doi = "10.1109/IEEECONF60004.2024.10942901",

language = "English",

series = "Conference Record - Asilomar Conference on Signals, Systems and Computers",

publisher = "IEEE",

pages = "1354--1359",

editor = "Matthews, {Michael B.}",

booktitle = "Conference Record of the 58th Asilomar Conference on Signals, Systems and Computers, ACSSC 2024",

address = "United States",

}

Chen, L, Hassan, M, Ravindran, S & Wu, R 2024, Track-Cued Radar Point Cloud Target Classification. in MB Matthews (ed.), Conference Record of the 58th Asilomar Conference on Signals, Systems and Computers, ACSSC 2024. Conference Record - Asilomar Conference on Signals, Systems and Computers, IEEE, pp. 1354-1359, 58th Asilomar Conference on Signals, Systems and Computers, ACSSC 2024, Hybrid, Pacific Grove, United States, 27/10/24. https://doi.org/10.1109/IEEECONF60004.2024.10942901

Track-Cued Radar Point Cloud Target Classification. / Chen, Lihui; Hassan, Mujtaba; Ravindran, Satish et al.
Conference Record of the 58th Asilomar Conference on Signals, Systems and Computers, ACSSC 2024. ed. / Michael B. Matthews. IEEE, 2024. p. 1354-1359 (Conference Record - Asilomar Conference on Signals, Systems and Computers).

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Track-Cued Radar Point Cloud Target Classification

AU - Chen, Lihui

AU - Hassan, Mujtaba

AU - Ravindran, Satish

AU - Wu, Ryan

N1 - Green Open Access added to TU Delft Institutional Repository ‘You share, we take care!’ – Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

PY - 2024

Y1 - 2024

N2 - A novel temporal-spatial object classification neural network model is proposed to improve the classification capability of tracked objects. It takes queued points of tracked objects using multiple frames as input, utilizes spatial and temporal information from these points for sampling and grouping as well as extracts hierarchical temporal-spatial features for target classification. Experimental results on a proprietary 4D Imaging Radar dataset and open-source 2D RadarScenes dataset demonstrate that the proposed tracker-cued radar point-cloud target classification method allows the model to learn meaningful appearance and motion features from sparse radar points data, and achieves accurate classification output as compared to a baseline method, while being efficient to run on edge hardware with limited resources.

AB - A novel temporal-spatial object classification neural network model is proposed to improve the classification capability of tracked objects. It takes queued points of tracked objects using multiple frames as input, utilizes spatial and temporal information from these points for sampling and grouping as well as extracts hierarchical temporal-spatial features for target classification. Experimental results on a proprietary 4D Imaging Radar dataset and open-source 2D RadarScenes dataset demonstrate that the proposed tracker-cued radar point-cloud target classification method allows the model to learn meaningful appearance and motion features from sparse radar points data, and achieves accurate classification output as compared to a baseline method, while being efficient to run on edge hardware with limited resources.

KW - Object classification

KW - temporal-spatial feature extraction

KW - temporal-spatial grouping

KW - temporal-spatial sampling

UR - http://www.scopus.com/inward/record.url?scp=105002687627&partnerID=8YFLogxK

U2 - 10.1109/IEEECONF60004.2024.10942901

DO - 10.1109/IEEECONF60004.2024.10942901

M3 - Conference contribution

AN - SCOPUS:105002687627

T3 - Conference Record - Asilomar Conference on Signals, Systems and Computers

SP - 1354

EP - 1359

BT - Conference Record of the 58th Asilomar Conference on Signals, Systems and Computers, ACSSC 2024

A2 - Matthews, Michael B.

PB - IEEE

T2 - 58th Asilomar Conference on Signals, Systems and Computers, ACSSC 2024

Y2 - 27 October 2024 through 30 October 2024

ER -

Track-Cued Radar Point Cloud Target Classification

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this