Mcrood: Multi-Class Radar Out-Of-Distribution Detection

Sabri Mustafa Kahya, Muhammet Sami Yavuz, Eckehard Steinbach

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

2 Zitate (Scopus)

Abstract

Out-of-distribution (OOD) detection has recently received special attention due to its critical role in safely deploying modern deep learning (DL) architectures. This work proposes a reconstruction-based multi-class OOD detector that operates on radar range doppler images (RDIs). The detector aims to classify any moving object other than a person sitting, standing, or walking as OOD. We also provide a simple yet effective pre-processing technique to detect minor human body movements like breathing. The simple idea is called respiration detector (RESPD) and eases the OOD detection, especially for human sitting and standing classes. On our dataset collected by 60GHz short-range FMCW Radar, we achieve AUROCs of 97.45%, 92.13%, and 96.58% for sitting, standing, and walking classes, respectively. We perform extensive experiments and show that our method outperforms state-of-the-art (SOTA) OOD detection methods. Also, our pipeline performs 24 times faster than the second-best method and is very suitable for real-time processing.

OriginalspracheEnglisch
TitelICASSP 2023 - 2023 IEEE International Conference on Acoustics, Speech and Signal Processing, Proceedings
Herausgeber (Verlag)Institute of Electrical and Electronics Engineers Inc.
ISBN (elektronisch)9781728163277
DOIs
PublikationsstatusVeröffentlicht - 2023
Veranstaltung48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023 - Rhodes Island, Griechenland
Dauer: 4 Juni 202310 Juni 2023

Publikationsreihe

NameICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Band2023-June
ISSN (Print)1520-6149

Konferenz

Konferenz48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023
Land/GebietGriechenland
OrtRhodes Island
Zeitraum4/06/2310/06/23

Fingerprint

Untersuchen Sie die Forschungsthemen von „Mcrood: Multi-Class Radar Out-Of-Distribution Detection“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren