Harnessing Unlabeled Data to Improve Generalization of Biometric Gender and Age Classifiers

Nadimpalli, Aakash Varma; Reddy, Narsi; Ramachandran, Sreeraj; Rattani, Ajita

doi:10.1109/ssci50451.2021.9660182

Cited by 6 publications

(2 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This draws attention to fairness and bias in AI-based facial analytics where unintended consequences from biased systems call for a thorough examination of the datasets and models [18,5,17,8,4]. Most of the published research in this domain suggests low performance for women, and dark-skinned people for facial attribute-based classification systems such as gender and age [8,17,30,24], and face recognition [5,4]. As biased datasets produce biased models, many of the efforts have been focused on developing gender and race-balanced datasets for various facial-analysis based applications.…”

Section: Celebdf Distributionmentioning

confidence: 99%

GBDF: Gender Balanced DeepFake Dataset Towards Fair DeepFake Detection

Nadimpalli¹,

Rattani²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Facial forgery by deepfakes has raised severe societal concerns. Several solutions have been proposed by the vision community to effectively combat the misinformation on the internet via automated deepfake detection systems. Recent studies have demonstrated that facial analysis-based deep learning models can discriminate based on protected attributes. For the commercial adoption and massive roll-out of the deepfake detection technology, it is vital to evaluate and understand the fairness (the absence of any prejudice or favoritism) of deepfake detectors across demographic variations such as gender and race. As the performance differential of deepfake detectors between demographic subgroups would impact millions of people of the deprived sub-group. This paper aims to evaluate the fairness of the deepfake detectors across males and females. However, existing deepfake datasets are not annotated with demographic labels to facilitate fairness analysis. To this aim, we manually annotated existing popular deepfake datasets with gender labels and evaluated the performance differential of current deepfake detectors across gender. Our analysis on the gender-labeled version of the datasets suggests (a) current deepfake datasets have skewed distribution across gender, and (b) commonly adopted deepfake detectors obtain unequal performance across gender with mostly males outperforming females. Finally, we contributed a gender-balanced and annotated deepfake dataset, GBDF, to mitigate the performance differential and to promote research and development towards fairness-aware deep fake detectors. The GBDF dataset is publicly available at: https://github.com/aakash4305/GBDF

show abstract

Section: Celebdf Distributionmentioning

confidence: 99%

GBDF: Gender Balanced DeepFake Dataset Towards Fair DeepFake Detection

Nadimpalli¹,

Rattani²

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…"São Paulo subway ordered to suspend use of facial recognition" [19], this is one among the several media articles published at a well-known press that suggested facial recognition technology is biased across demographics. Apart from media articles, several published academic studies on the bias of face recognition and visual attribute classification algorithms (such as gender-, age-classification and BMI prediction) also suggest performance differences across gender, race, and age [4]- [6], [9], [14], [16], [17], [21], [22], [28]. An unfair (biased) algorithm is one whose decisions are skewed towards a particular group of people.…”

Section: Introductionmentioning

confidence: 99%

Is Facial Recognition Biased at Near-Infrared Spectrum As Well?

Krishnan¹,

Neas²,

Rattani³

2022

Preprint

View full text Add to dashboard Cite

Published academic research and media articles suggest face recognition is biased across demographics. Specifically, unequal performance is obtained for women, dark-skinned people, and older adults. However, these published studies have examined the bias of facial recognition in the visible spectrum (VIS). Factors such as facial makeup, facial hair, skin color, and illumination variation have been attributed to the bias of this technology at VIS. The near-infrared (NIR) spectrum offers an advantage over VIS in terms of robustness to factors such as illumination changes, facial make-up, and skin color.Therefore, it is worth-while to investigate the bias of the facial recognition at near-infrared spectrum (NIR). This first study investigates the bias of face recognition system at NIR spectrum. To this aim, two popular NIR facial image datasets namely, CASIA-Face-Africa and NotreDame-NIVL consisting of African and Caucasian subjects, respectively, are used to investigate the bias of facial recognition technology across gender and race. Interestingly, experimental results suggest equitable performance of the face recognition across gender and race at NIR spectrum.

show abstract

GBDF: Gender Balanced DeepFake Dataset Towards Fair DeepFake Detection

Nadimpalli

Rattani

2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Harnessing Unlabeled Data to Improve Generalization of Biometric Gender and Age Classifiers

Cited by 6 publications

References 23 publications

GBDF: Gender Balanced DeepFake Dataset Towards Fair DeepFake Detection

GBDF: Gender Balanced DeepFake Dataset Towards Fair DeepFake Detection

Is Facial Recognition Biased at Near-Infrared Spectrum As Well?

GBDF: Gender Balanced DeepFake Dataset Towards Fair DeepFake Detection

Contact Info

Product

Resources

About