Skip to main content

Research Repository

Advanced Search

Race Bias Analysis of Bona Fide Errors in Face Anti-spoofing

Abduh, Latifah; Ivrissimtzis, Ioannis

Race Bias Analysis of Bona Fide Errors in Face Anti-spoofing Thumbnail


Authors

Latifah Abduh latifah.a.abduh@durham.ac.uk
PGR Student Doctor of Philosophy



Abstract

The study of bias in Machine Learning is receiving a lot of attention in recent years, however, few only papers deal explicitly with the problem of race bias in face anti-spoofing. In this paper, we present a systematic study of race bias in face anti-spoofing with three key features: we focus on the classifier’s bona fide errors, where the most significant ethical and legal issues lie; we analyse both the scalar responses of the classifier and its final binary outcomes; the threshold determining the operating point of the classifier is treated as a variable. We apply the proposed bias analysis framework on a VQ-VAE-based face anti-spoofing algorithm. Our main conclusion is that race bias should not necessarily be attributed to different mean values of the response distributions over the various demographics. Instead, it can be better understood as the combined effect of several possible characteristics of these distributions: different means; different variances; bimodal behaviour; the existence of outliers.

Citation

Abduh, L., & Ivrissimtzis, I. (2023, September). Race Bias Analysis of Bona Fide Errors in Face Anti-spoofing. Presented at CAIP 2023: The 20th International Conference on Computer Analysis of Images and Patterns, Limassol, Cyprus

Presentation Conference Type Conference Paper (Published)
Conference Name CAIP 2023: The 20th International Conference on Computer Analysis of Images and Patterns
Start Date Sep 25, 2023
End Date Sep 28, 2023
Acceptance Date Jun 30, 2023
Online Publication Date Sep 20, 2023
Publication Date 2023
Deposit Date Aug 23, 2023
Publicly Available Date Sep 21, 2024
Print ISSN 0302-9743
Publisher Springer
Volume 14185
Pages 23-32
Series Title Lecture Notes in Computer Science
Series ISSN 0302-9743
Book Title CAIP 2023: Computer Analysis of Images and Patterns
ISBN 9783031442391
DOI https://doi.org/10.1007/978-3-031-44240-7_3
Public URL https://durham-repository.worktribe.com/output/1723511
Publisher URL https://link.springer.com/conference/caip

Files





You might also like



Downloadable Citations