Does lossy image compression affect racial bias within face recognition?

TL;DR

This paper examines how lossy image compression, especially JPEG, impacts face recognition accuracy across different racial phenotypes, revealing increased bias against darker skin tones and the effects of chroma-subsampling.

Contribution

It introduces a racial phenotype-based bias analysis methodology and explores how chroma-subsampling influences race-related recognition performance disparities.

Findings

Lossy compression worsens recognition accuracy for darker skin tones by up to 34.55%.

Removing chroma-subsampling reduces false matches by up to 15.95%.

Chroma-subsampling characteristics contribute to racial bias in face recognition.

Abstract

Yes - This study investigates the impact of commonplace lossy image compression on face recognition algorithms with regard to the racial characteristics of the subject. We adopt a recently proposed racial phenotype-based bias analysis methodology to measure the effect of varying levels of lossy compression across racial phenotype categories. Additionally, we determine the relationship between chroma-subsampling and race-related phenotypes for recognition performance. Prior work investigates the impact of lossy JPEG compression algorithm on contemporary face recognition performance. However, there is a gap in how this impact varies with different race-related inter-sectional groups and the cause of this impact. Via an extensive experimental setup, we demonstrate that common lossy image compression approaches have a more pronounced negative impact on facial recognition performance for specific racial phenotype categories such as darker skin tones (by up to 34.55\%). Furthermore, removing chroma-subsampling during compression improves the false matching rate (up to 15.95\%) across all phenotype categories affected by the compression, including darker skin tones, wide noses, big lips, and monolid eye categories. In addition, we outline the characteristics that may be attributable as the underlying cause of such phenomenon for lossy compression algorithms such as JPEG.