Analysis of Gender Inequality In Face Recognition Accuracy

Abstract

We present a comprehensive analysis of how and why face recognition accuracy differs between men and women. We show that accuracy is lower for women due to the combination of (1) the impostor distribution for women having a skew toward higher similarity scores, and (2) the genuine distribution for women having a skew toward lower similarity scores. We show that this phenomenon of the impostor and genuine distributions for women shifting closer towards each other is general across datasets of African-American, Caucasian, and Asian faces. We show that the distribution of facial expressions may differ between male/female, but that the accuracy difference persists for image subsets rated confidently as neutral expression. The accuracy difference also persists for image subsets rated as close to zero pitch angle. Even when removing images with forehead partially occluded by hair/hat, the same impostor/genuine accuracy difference persists. We show that the female genuine distribution improves when only female images without facial cosmetics are used, but that the female impostor distribution also degrades at the same time. Lastly, we show that the accuracy difference persists even if a state-of-the-art deep learning method is trained from scratch using training data explicitly balanced between male and female images and subjects.

Towards Causal Benchmarking of Bias in Face Analysis Algorithms

“…Computer vision datasets are often found to be biased [64,76]. Human face datasets are particularly scrutinized [2,20,43,45,46,54] because methods and models trained on these data can end up being biased along attributes that are protected by the law [44]. Approaches to mitigating dataset bias include collecting more thorough examples [54], using image synthesis to compensate for distribution gaps [46], and example resampling [48].…”

“…Recent work uses generative models to explore face classification system biases. One study explores how variations in pose and lighting affect classifier performance [2,45,46]. A second study uses a generative model to synthesize faces along particular attribute directions [19].…”

Collection of 2429 constrained headshots of 277 volunteers for deep learning

“…Several studies have concluded that face recognition is more accurate in men than in women (Buolamwini and Gebru, 2018;Albiero et al, 2020). Accordingly, our CNN committed errors in 2 male and 3 female cases.…”

Golden Ratio and Its Application to Bayes Classifier Based Face Sketch Gender Classification and Recognition

“…Face Gender Recognition (FGR) system is a major area for non-verbal language in day to day life communication. FGR systems have been attracted numerous researchers since they attempt to overcome the problems and factors weakening these systems including problem of images classification, also due to its large-scale applications in face analysis, particularly face recognition [1]. Gender based separation among humans is classified into two: male and female [2].…”

“…It based on two-dimensional images of human subjects. Currently gender classification and recognition from facial imagery has grown its importance in the computer vision field: It play a very important function in many fields likes, face recognition [1][3], forensic crime detection [4] [5], facial emotion recognition [3] and psychologically affected patients [6], night surveillance [7] and Artificial Intelligence [8] [9] and soon. In this paper it can be used to identify; fastly; a criminal person from his sketch for purposes of identification [10].…”