Real-Time Multi-object Tracking and Identification Using Sparse Point-Cloud Data from Low-Cost mmWave Radar

Nabih Pico; Maykoll Vanegas; Eugene Auh; Hong Ryul Jung; Altair Coutinho; Elvia Montero; Hyungpil Moon

doi:10.1007/978-3-031-92011-0_12

Real-Time Multi-object Tracking and Identification Using Sparse Point-Cloud Data from Low-Cost mmWave Radar

Nabih Pico
, Maykoll Vanegas
, Eugene Auh
, Hong Ryul Jung
, Altair Coutinho
, Elvia Montero
, Hyungpil Moon

Department of Mechanical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This paper proposes an efficient, real-time method for recognizing and tracking multiple objects using sparse point-cloud sequences generated by a low-cost mmWave radar. The system employs the DBSCAN algorithm to cluster the radar’s point cloud data, capturing objects across multiple frames within defined windows. A moving average filter is applied to mitigate measurement errors in the radar data. During the tracking phase, a Kalman filter predicts object positions, while the Hungarian algorithm ensures the correct assignment of detections to specific tracks. The proposed method is evaluated through five experiments, where people move within the radar’s field of view. These experiments involve overlapping people, making the tracking algorithm particularly challenging. The Multi-Object Tracking Accuracy (MOTA) metric is used to assess the results, achieving a 90.10% accuracy rate, which underscores the method’s potential for real-time multi-object tracking using mmWave radar. Videos of the experiments can be accessed via the following link: https://github.com/nabihandres/RADAR_tracking_tests.git

Original language	English
Title of host publication	Robot Intelligence Technology and Applications 9 - Results from the 12th International Conference on Robot Intelligence Technology and Applications
Editors	Daehyung Park, Dae-Young Lee, Min Jun Kim, Cunjia Liu
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	143-154
Number of pages	12
ISBN (Print)	9783031920103
DOIs	https://doi.org/10.1007/978-3-031-92011-0_12
State	Published - 2025
Event	12th International Conference on Robot Intelligence Technology and Applications, RiTA 2024 - Ulsan, Korea, Republic of Duration: 4 Dec 2024 → 7 Dec 2024

Publication series

Name	Lecture Notes in Networks and Systems
Volume	1419 LNNS
ISSN (Print)	2367-3370
ISSN (Electronic)	2367-3389

Conference

Conference	12th International Conference on Robot Intelligence Technology and Applications, RiTA 2024
Country/Territory	Korea, Republic of
City	Ulsan
Period	4/12/24 → 7/12/24

Keywords

DBSCAN
Kalman Filter
mmWave Radar
Multi-Object Tracking

Access to Document

10.1007/978-3-031-92011-0_12

Cite this

Pico, N., Vanegas, M., Auh, E., Jung, H. R., Coutinho, A., Montero, E., & Moon, H. (2025). Real-Time Multi-object Tracking and Identification Using Sparse Point-Cloud Data from Low-Cost mmWave Radar. In D. Park, D.-Y. Lee, M. J. Kim, & C. Liu (Eds.), Robot Intelligence Technology and Applications 9 - Results from the 12th International Conference on Robot Intelligence Technology and Applications (pp. 143-154). (Lecture Notes in Networks and Systems; Vol. 1419 LNNS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-92011-0_12

Pico, Nabih ; Vanegas, Maykoll ; Auh, Eugene et al. / Real-Time Multi-object Tracking and Identification Using Sparse Point-Cloud Data from Low-Cost mmWave Radar. Robot Intelligence Technology and Applications 9 - Results from the 12th International Conference on Robot Intelligence Technology and Applications. editor / Daehyung Park ; Dae-Young Lee ; Min Jun Kim ; Cunjia Liu. Springer Science and Business Media Deutschland GmbH, 2025. pp. 143-154 (Lecture Notes in Networks and Systems).

@inproceedings{c0c6cf117f1c460fb9b5d73b5ebe1728,

title = "Real-Time Multi-object Tracking and Identification Using Sparse Point-Cloud Data from Low-Cost mmWave Radar",

abstract = "This paper proposes an efficient, real-time method for recognizing and tracking multiple objects using sparse point-cloud sequences generated by a low-cost mmWave radar. The system employs the DBSCAN algorithm to cluster the radar{\textquoteright}s point cloud data, capturing objects across multiple frames within defined windows. A moving average filter is applied to mitigate measurement errors in the radar data. During the tracking phase, a Kalman filter predicts object positions, while the Hungarian algorithm ensures the correct assignment of detections to specific tracks. The proposed method is evaluated through five experiments, where people move within the radar{\textquoteright}s field of view. These experiments involve overlapping people, making the tracking algorithm particularly challenging. The Multi-Object Tracking Accuracy (MOTA) metric is used to assess the results, achieving a 90.10\% accuracy rate, which underscores the method{\textquoteright}s potential for real-time multi-object tracking using mmWave radar. Videos of the experiments can be accessed via the following link: https://github.com/nabihandres/RADAR\_tracking\_tests.git",

keywords = "DBSCAN, Kalman Filter, mmWave Radar, Multi-Object Tracking",

author = "Nabih Pico and Maykoll Vanegas and Eugene Auh and Jung, \{Hong Ryul\} and Altair Coutinho and Elvia Montero and Hyungpil Moon",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.; 12th International Conference on Robot Intelligence Technology and Applications, RiTA 2024 ; Conference date: 04-12-2024 Through 07-12-2024",

year = "2025",

doi = "10.1007/978-3-031-92011-0\_12",

language = "English",

isbn = "9783031920103",

series = "Lecture Notes in Networks and Systems",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "143--154",

editor = "Daehyung Park and Dae-Young Lee and Kim, \{Min Jun\} and Cunjia Liu",

booktitle = "Robot Intelligence Technology and Applications 9 - Results from the 12th International Conference on Robot Intelligence Technology and Applications",

}

Pico, N, Vanegas, M, Auh, E, Jung, HR, Coutinho, A, Montero, E & Moon, H 2025, Real-Time Multi-object Tracking and Identification Using Sparse Point-Cloud Data from Low-Cost mmWave Radar. in D Park, D-Y Lee, MJ Kim & C Liu (eds), Robot Intelligence Technology and Applications 9 - Results from the 12th International Conference on Robot Intelligence Technology and Applications. Lecture Notes in Networks and Systems, vol. 1419 LNNS, Springer Science and Business Media Deutschland GmbH, pp. 143-154, 12th International Conference on Robot Intelligence Technology and Applications, RiTA 2024, Ulsan, Korea, Republic of, 4/12/24. https://doi.org/10.1007/978-3-031-92011-0_12

Real-Time Multi-object Tracking and Identification Using Sparse Point-Cloud Data from Low-Cost mmWave Radar. / Pico, Nabih; Vanegas, Maykoll; Auh, Eugene et al.
Robot Intelligence Technology and Applications 9 - Results from the 12th International Conference on Robot Intelligence Technology and Applications. ed. / Daehyung Park; Dae-Young Lee; Min Jun Kim; Cunjia Liu. Springer Science and Business Media Deutschland GmbH, 2025. p. 143-154 (Lecture Notes in Networks and Systems; Vol. 1419 LNNS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Real-Time Multi-object Tracking and Identification Using Sparse Point-Cloud Data from Low-Cost mmWave Radar

AU - Pico, Nabih

AU - Vanegas, Maykoll

AU - Auh, Eugene

AU - Jung, Hong Ryul

AU - Coutinho, Altair

AU - Montero, Elvia

AU - Moon, Hyungpil

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

PY - 2025

Y1 - 2025

N2 - This paper proposes an efficient, real-time method for recognizing and tracking multiple objects using sparse point-cloud sequences generated by a low-cost mmWave radar. The system employs the DBSCAN algorithm to cluster the radar’s point cloud data, capturing objects across multiple frames within defined windows. A moving average filter is applied to mitigate measurement errors in the radar data. During the tracking phase, a Kalman filter predicts object positions, while the Hungarian algorithm ensures the correct assignment of detections to specific tracks. The proposed method is evaluated through five experiments, where people move within the radar’s field of view. These experiments involve overlapping people, making the tracking algorithm particularly challenging. The Multi-Object Tracking Accuracy (MOTA) metric is used to assess the results, achieving a 90.10% accuracy rate, which underscores the method’s potential for real-time multi-object tracking using mmWave radar. Videos of the experiments can be accessed via the following link: https://github.com/nabihandres/RADAR_tracking_tests.git

AB - This paper proposes an efficient, real-time method for recognizing and tracking multiple objects using sparse point-cloud sequences generated by a low-cost mmWave radar. The system employs the DBSCAN algorithm to cluster the radar’s point cloud data, capturing objects across multiple frames within defined windows. A moving average filter is applied to mitigate measurement errors in the radar data. During the tracking phase, a Kalman filter predicts object positions, while the Hungarian algorithm ensures the correct assignment of detections to specific tracks. The proposed method is evaluated through five experiments, where people move within the radar’s field of view. These experiments involve overlapping people, making the tracking algorithm particularly challenging. The Multi-Object Tracking Accuracy (MOTA) metric is used to assess the results, achieving a 90.10% accuracy rate, which underscores the method’s potential for real-time multi-object tracking using mmWave radar. Videos of the experiments can be accessed via the following link: https://github.com/nabihandres/RADAR_tracking_tests.git

KW - DBSCAN

KW - Kalman Filter

KW - mmWave Radar

KW - Multi-Object Tracking

UR - https://www.scopus.com/pages/publications/105011942666

U2 - 10.1007/978-3-031-92011-0_12

DO - 10.1007/978-3-031-92011-0_12

M3 - Conference contribution

AN - SCOPUS:105011942666

SN - 9783031920103

T3 - Lecture Notes in Networks and Systems

SP - 143

EP - 154

BT - Robot Intelligence Technology and Applications 9 - Results from the 12th International Conference on Robot Intelligence Technology and Applications

A2 - Park, Daehyung

A2 - Lee, Dae-Young

A2 - Kim, Min Jun

A2 - Liu, Cunjia

PB - Springer Science and Business Media Deutschland GmbH

T2 - 12th International Conference on Robot Intelligence Technology and Applications, RiTA 2024

Y2 - 4 December 2024 through 7 December 2024

ER -

Pico N, Vanegas M, Auh E, Jung HR, Coutinho A, Montero E et al. Real-Time Multi-object Tracking and Identification Using Sparse Point-Cloud Data from Low-Cost mmWave Radar. In Park D, Lee DY, Kim MJ, Liu C, editors, Robot Intelligence Technology and Applications 9 - Results from the 12th International Conference on Robot Intelligence Technology and Applications. Springer Science and Business Media Deutschland GmbH. 2025. p. 143-154. (Lecture Notes in Networks and Systems). doi: 10.1007/978-3-031-92011-0_12