Skip to main navigation Skip to search Skip to main content

Dual-Timescale Classification of Human Activities Using Radar Point Clouds

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

Abstract

The problem of radar-based, continuous Human Activity Recognition (HAR) has been studied in this work. A fixed-window segmentation method based on dual timescales has been proposed to tackle this challenge. The method is experimentally validated on a challenging publicly available dataset with 14 participants and 9 activities, and is compared to reference works from the literature. L1PO validation of the method yields a test accuracy and macro F1-score of 87.5 % and 80.1 % respectively.
Original languageEnglish
Title of host publicationProceedings of the 2025 22nd European Radar Conference (EuRAD)
PublisherIEEE
Pages111-114
Number of pages4
ISBN (Electronic)978-2-87487-083-5
ISBN (Print)979-8-3315-3649-7
DOIs
Publication statusPublished - 2025
Event2025 22nd European Radar Conference (EuRAD) - Utrecht, Netherlands
Duration: 24 Sept 202526 Sept 2025

Publication series

Name2025 22nd European Radar Conference, EuRAD 2025

Conference

Conference2025 22nd European Radar Conference (EuRAD)
Country/TerritoryNetherlands
CityUtrecht
Period24/09/2526/09/25

Bibliographical note

Green Open Access added to TU Delft Institutional Repository as part of the Taverne amendment. More information about this copyright law amendment can be found at https://www.openaccess.nl. 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

  • Human activity recognition
  • machine learning
  • radar
  • point cloud processing

Fingerprint

Dive into the research topics of 'Dual-Timescale Classification of Human Activities Using Radar Point Clouds'. Together they form a unique fingerprint.

Cite this