An Examination of Fairness of AI Models for Deepfake Detection

• URL: http://arxiv.org/abs/2105.00558v1
• Date: Sun, 2 May 2021 21:55:04 GMT
• Title: An Examination of Fairness of AI Models for Deepfake Detection
• Authors: Loc Trinh, Yan Liu
• Abstract summary: We evaluate bias present in deepfake datasets and detection models across protected subgroups. Using facial datasets balanced by race and gender, we examine three popular deepfake detectors and find large disparities in predictive performances across races.
• Score: 5.4852920337961235
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent studies have demonstrated that deep learning models can discriminate based on protected classes like race and gender. In this work, we evaluate bias present in deepfake datasets and detection models across protected subgroups. Using facial datasets balanced by race and gender, we examine three popular deepfake detectors and find large disparities in predictive performances across races, with up to 10.7% difference in error rate between subgroups. A closer look reveals that the widely used FaceForensics++ dataset is overwhelmingly composed of Caucasian subjects, with the majority being female Caucasians. Our investigation of the racial distribution of deepfakes reveals that the methods used to create deepfakes as positive training signals tend to produce "irregular" faces - when a person's face is swapped onto another person of a different race or gender. This causes detectors to learn spurious correlations between the foreground faces and fakeness. Moreover, when detectors are trained with the Blended Image (BI) dataset from Face X-Rays, we find that those detectors develop systematic discrimination towards certain racial subgroups, primarily female Asians.

Related papers

• From Bias to Balance: Detecting Facial Expression Recognition Biases in Large Multimodal Foundation Models [3.1927733045184885]
  This study addresses the racial biases in facial expression recognition (FER) systems within Large Multimodal Foundation Models (LMFMs) Existing research predominantly focuses on traditional FER models (CNNs, RNNs, ViTs) leaving a gap in understanding racial biases in LMFMs. We benchmark four leading LMFMs: GPT-4o, PaliGemma, Gemini, and CLIP to assess their performance in facial emotion detection across different racial demographics.
  arXiv  Detail & Related papers  (2024-08-27T07:54:01Z)
• Towards Fair Face Verification: An In-depth Analysis of Demographic Biases [11.191375513738361]
  Deep learning-based person identification and verification systems have remarkably improved in terms of accuracy in recent years. However, such systems have been found to exhibit significant biases related to race, age, and gender. This paper presents an in-depth analysis, with a particular emphasis on the intersectionality of these demographic factors.
  arXiv  Detail & Related papers  (2023-07-19T14:49:14Z)
• Are Face Detection Models Biased? [69.68854430664399]
  We investigate possible bias in the domain of face detection through facial region localization. Most existing face detection datasets lack suitable annotation for such analysis. We observe a high disparity in detection accuracies across gender and skin-tone, and interplay of confounding factors beyond demography.
  arXiv  Detail & Related papers  (2022-11-07T14:27:55Z)
• Mitigating Gender Bias in Face Recognition Using the von Mises-Fisher Mixture Model [7.049738935364298]
  In this work, we investigate the gender bias of deep Face Recognition networks. Motivated by geometric considerations, we mitigate gender bias through a new post-processing methodology. In fact, extensive numerical experiments on a variety of datasets show that a careful selection significantly reduces gender bias.
  arXiv  Detail & Related papers  (2022-10-24T23:53:56Z)
• Rethinking Bias Mitigation: Fairer Architectures Make for Fairer Face Recognition [107.58227666024791]
  Face recognition systems are widely deployed in safety-critical applications, including law enforcement. They exhibit bias across a range of socio-demographic dimensions, such as gender and race. Previous works on bias mitigation largely focused on pre-processing the training data.
  arXiv  Detail & Related papers  (2022-10-18T15:46:05Z)
• GBDF: Gender Balanced DeepFake Dataset Towards Fair DeepFake Detection [1.0152838128195467]
  Facial forgery by deepfakes has raised severe societal concerns. Recent studies have demonstrated that facial analysis-based deep learning models can discriminate based on protected attributes. It is vital to evaluate and understand the fairness of deepfake detectors across demographic variations such as gender and race.
  arXiv  Detail & Related papers  (2022-07-21T01:00:40Z)
• A Deep Dive into Dataset Imbalance and Bias in Face Identification [49.210042420757894]
  Media portrayals often center imbalance as the main source of bias in automated face recognition systems. Previous studies of data imbalance in FR have focused exclusively on the face verification setting. This work thoroughly explores the effects of each kind of imbalance possible in face identification, and discuss other factors which may impact bias in this setting.
  arXiv  Detail & Related papers  (2022-03-15T20:23:13Z)
• Are Commercial Face Detection Models as Biased as Academic Models? [64.71318433419636]
  We compare academic and commercial face detection systems, specifically examining robustness to noise. We find that state-of-the-art academic face detection models exhibit demographic disparities in their noise robustness. We conclude that commercial models are always as biased or more biased than an academic model.
  arXiv  Detail & Related papers  (2022-01-25T02:21:42Z)
• Balancing Biases and Preserving Privacy on Balanced Faces in the Wild [50.915684171879036]
  There are demographic biases present in current facial recognition (FR) models. We introduce our Balanced Faces in the Wild dataset to measure these biases across different ethnic and gender subgroups. We find that relying on a single score threshold to differentiate between genuine and imposters sample pairs leads to suboptimal results. We propose a novel domain adaptation learning scheme that uses facial features extracted from state-of-the-art neural networks.
  arXiv  Detail & Related papers  (2021-03-16T15:05:49Z)
• Understanding Fairness of Gender Classification Algorithms Across Gender-Race Groups [0.8594140167290097]
  The aim of this paper is to investigate the differential performance of the gender classification algorithms across gender-race groups. For all the algorithms used, Black females (Black race in general) always obtained the least accuracy rates. Middle Eastern males and Latino females obtained higher accuracy rates most of the time.
  arXiv  Detail & Related papers  (2020-09-24T04:56:10Z)
• Investigating Bias in Deep Face Analysis: The KANFace Dataset and Empirical Study [67.3961439193994]
  We introduce the most comprehensive, large-scale dataset of facial images and videos to date. The data are manually annotated in terms of identity, exact age, gender and kinship. A method to debias network embeddings is introduced and tested on the proposed benchmarks.
  arXiv  Detail & Related papers  (2020-05-15T00:14:39Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.