Does Face Recognition Error Echo Gender Classification Error?

TL;DR

This study investigates whether errors in gender classification influence face recognition errors, revealing that gender misclassification correlates with both reduced false matches in impostor pairs and increased false non-matches in genuine pairs.

Contribution

It is the first to analyze the relationship between gender classification errors and face recognition errors across multiple algorithms and demographic groups.

Findings

Gender errors reduce false match rates in impostor pairs.

Gender errors increase false non-match rates in genuine pairs.

Different error patterns are observed depending on whether images have correct or incorrect gender classification.

Abstract

This paper is the first to explore the question of whether images that are classified incorrectly by a face analytics algorithm (e.g., gender classification) are any more or less likely to participate in an image pair that results in a face recognition error. We analyze results from three different gender classification algorithms (one open-source and two commercial), and two face recognition algorithms (one open-source and one commercial), on image sets representing four demographic groups (African-American female and male, Caucasian female and male). For impostor image pairs, our results show that pairs in which one image has a gender classification error have a better impostor distribution than pairs in which both images have correct gender classification, and so are less likely to generate a false match error. For genuine image pairs, our results show that individuals whose images have a mix of correct and incorrect gender classification have a worse genuine distribution (increased false non-match rate) compared to individuals whose images all have correct gender classification. Thus, compared to images that generate correct gender classification, images that generate gender classification errors do generate a different pattern of recognition errors, both better (false match) and worse (false non-match).