RFOOD: Real-time Facial Authentication and Out -of-distribution Detection with Short-range FMCW Radar

Sabri Mustafa Kahya, Muhammet Sami Yavuz, Boran Hamdi Sivrikaya, Eckehard Steinbach

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Out-of-distribution (OOD) detection is critical for the safe deployment of modern neural network architectures, as it aims to identify samples outside the training domain. In this paper, we introduce RFOOD, a novel OOD detection framework designed for real-time, privacy-preserving facial authentication using low-cost frequency-modulated continuous-wave (FMCW) radar. RFOOD employs both range-Doppler and micro range- Doppler images to enhance the detection accuracy. The architecture consists of a multi-encoder multi-decoder Body Part (BP) and Intermediate Linear Encoder-Decoder (ILED) components. This design allows the system to accurately classify a single individual's face as in-distribution (ID) while identifying all other faces as OOD. On our dataset collected with 60 GHz short-range FMCW radar, RFOOD achieves an Area Under the Receiver Operating Characteristic (AUROC) curve of 94.13 % and a False Positive Rate of 18.12% at a True Positive Rate of 95 % (FPR95). Additionally, RFOOD outperforms state-of-the-art OOD detection methods in common OOD detection metrics and operates in real-time.

Original languageEnglish
Title of host publicationProceedings - 2024 International Conference on Machine Learning and Applications, ICMLA 2024
EditorsM. Arif Wani, Plamen Angelov, Feng Luo, Mitsunori Ogihara, Xintao Wu, Radu-Emil Precup, Ramin Ramezani, Xiaowei Gu
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages528-533
Number of pages6
ISBN (Electronic)9798350374889
DOIs
StatePublished - 2024
Event23rd IEEE International Conference on Machine Learning and Applications, ICMLA 2024 - Miami, United States
Duration: 18 Dec 202420 Dec 2024

Publication series

NameProceedings - 2024 International Conference on Machine Learning and Applications, ICMLA 2024

Conference

Conference23rd IEEE International Conference on Machine Learning and Applications, ICMLA 2024
Country/TerritoryUnited States
CityMiami
Period18/12/2420/12/24

Keywords

  • 60 GHz FMCW radar
  • deep neural networks
  • Facial authentication
  • out-of-distribution detection

Fingerprint

Dive into the research topics of 'RFOOD: Real-time Facial Authentication and Out -of-distribution Detection with Short-range FMCW Radar'. Together they form a unique fingerprint.

Cite this