Filters








166,100 Hits in 5.8 sec

Face Recognition: Too Bias, or Not Too Bias? [article]

Joseph P Robinson, Gennady Livitz, Yann Henon, Can Qin, Yun Fu, Samson Timoner
2020 arXiv   pre-print
We reveal critical insights into problems of bias in state-of-the-art facial recognition (FR) systems using a novel Balanced Faces In the Wild (BFW) dataset: data balanced for gender and ethnic groups.  ...  We show variations in the optimal scoring threshold for face-pairs across different subgroups.  ...  biased, or not?  ... 
arXiv:2002.06483v4 fatcat:464dhibgjvfzvaz2f3wlk2blgm

Face Recognition: Too Bias, or Not Too Bias?

Joseph P Robinson, Gennady Livitz, Yann Henon, Can Qin, Yun Fu, Samson Timoner
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
We reveal critical insights into problems of bias in state-of-the-art facial recognition (FR) systems using a novel Balanced Faces In the Wild (BFW) dataset: data balanced for gender and ethnic groups.  ...  We show variations in the optimal scoring threshold for face-pairs across different subgroups.  ...  That skew (i.e., the difference in the performance of an algorithm of particular demographics) is our definition of bias. A key question is: is FR too biased, or not?  ... 
doi:10.1109/cvprw50498.2020.00008 dblp:conf/cvpr/RobinsonLHQ0T20 fatcat:g5dkzhys2jcc7iasys3qsdxli4

Mitigating Face Recognition Bias via Group Adaptive Classifier [article]

Sixue Gong, Xiaoming Liu, Anil K. Jain
2020 arXiv   pre-print
Face recognition is known to exhibit bias - subjects in a certain demographic group can be better recognized than other groups.  ...  Experiments on face benchmarks (RFW, LFW, IJB-A, and IJB-C) show that our work is able to mitigate face recognition bias across demographic groups while maintaining the competitive accuracy.  ...  The imbalanced distribution of demographics in face data is, nevertheless, not the only trigger of FR bias.  ... 
arXiv:2006.07576v2 fatcat:kuhwbpz4mjcdflk5ce2mig7pmq

Distill and De-bias: Mitigating Bias in Face Verification using Knowledge Distillation [article]

Prithviraj Dhar, Joshua Gleason, Aniket Roy, Carlos D. Castillo, P. Jonathon Phillips, Rama Chellappa
2022 arXiv   pre-print
Face recognition networks generally demonstrate bias with respect to sensitive attributes like gender, skintone etc.  ...  We show that D&D++ outperforms existing baselines in reducing gender and skintone bias on the IJB-C dataset, while obtaining higher face verification performance than existing adversarial de-biasing methods  ...  Consistent [56] JP Robinson, G Livitz, Y Henon, C Qin, Y Fu, and S Timo- instance false positive improves fairness in face recognition. ner. Face recognition: too bias, or not too bias?  ... 
arXiv:2112.09786v3 fatcat:fi2ame4hlffnvhoxsauhfpaqkq

Exploring Bias in Primate Face Detection and Recognition [chapter]

Sanchit Sinha, Mohit Agarwal, Mayank Vatsa, Richa Singh, Saket Anand
2019 Lecture Notes in Computer Science  
It is worth exploring whether the knowledge of human faces and recent methods learned from human face detection and recognition can be extended to primate faces.  ...  However, similar challenges relating to bias in human faces will also occur in primates.  ...  For instance, face recognition of tigers or monkeys may not have prior literature or database as the starting point.  ... 
doi:10.1007/978-3-030-11009-3_33 fatcat:netdmywhs5hsnn7jx6zbmong3u

Comparing Human and Machine Bias in Face Recognition [article]

Samuel Dooley, Ryan Downing, George Wei, Nathan Shankar, Bradon Thymes, Gudrun Thorkelsdottir, Tiye Kurtz-Miott, Rachel Mattson, Olufemi Obiwumi, Valeriia Cherepanova, Micah Goldblum, John P Dickerson (+1 others)
2021 arXiv   pre-print
CelebA, and (2) do not compare their observed algorithmic bias to the biases of their human alternatives.  ...  These audits are immensely important and successful at measuring algorithmic bias but have two major challenges: the audits (1) use facial recognition datasets which lack quality metadata, like LFW and  ...  bias compares to the biases of machines or humanmachine teams.  ... 
arXiv:2110.08396v2 fatcat:nnzbmppyqndihab7ok3pekophu

Measuring Hidden Bias within Face Recognition via Racial Phenotypes [article]

Seyma Yucer, Furkan Tektas, Noura Al Moubayed, Toby P. Breckon
2021 arXiv   pre-print
We propose categorical test cases to investigate the individual influence of those attributes on bias within face recognition tasks.  ...  By contrast, this study introduces an alternative racial bias analysis methodology via facial phenotype attributes for face recognition.  ...  There are two scenarios for face identification applications based on whether a queried face is enrolled in a database or not.  ... 
arXiv:2110.09839v1 fatcat:5lz62g5i7ncy7bdgff33gpndhi

FairFace Challenge at ECCV 2020: Analyzing Bias in Face Recognition [article]

Tomáš Sixta, Julio C. S. Jacques Junior, Pau Buch-Cardona, Neil M. Robertson, Eduard Vazquez, Sergio Escalera
2020 arXiv   pre-print
This work summarizes the 2020 ChaLearn Looking at People Fair Face Recognition and Analysis Challenge and provides a description of the top-winning solutions and analysis of the results.  ...  Common strategies by the participants were face pre-processing, homogenization of data distributions, the use of bias aware loss functions and ensemble models.  ...  Bias Analysis In this section we analyze biases in the results of top-10 teams and discuss their possible causes.  ... 
arXiv:2009.07838v2 fatcat:2mty6zmyjjb77an2wnbdemppvm

Measuring Hidden Bias within Face Recognition via Racial Phenotypes

Seyma Yucer, Furkan Tektas, Noura Al Moubayed, Toby P. Breckon
2022 2022 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)  
We propose categorical test cases to investigate the individual influence of those attributes on bias within face recognition tasks.  ...  By contrast, this study introduces an alternative racial bias analysis methodology via facial phenotype attributes for face recognition.  ...  There are two scenarios for face identification applications based on whether a queried face is enrolled in a database or not.  ... 
doi:10.1109/wacv51458.2022.00326 fatcat:my4vfzf6ozcjjgo5lcy225hqqe

PASS: Protected Attribute Suppression System for Mitigating Bias in Face Recognition [article]

Prithviraj Dhar, Joshua Gleason, Aniket Roy, Carlos D. Castillo, Rama Chellappa
2021 arXiv   pre-print
Such encoding has two major issues: (a) it makes the face representations susceptible to privacy leakage (b) it appears to contribute to bias in face recognition.  ...  Face recognition networks encode information about sensitive attributes while being trained for identity classification.  ...  In Proceedings of the European conference on Face recognition: too bias, or not too bias?  ... 
arXiv:2108.03764v1 fatcat:eiehasmayrfnlfdxjzi7wgqu3y

Short article: The effects of precedence on Navon-induced processing bias in face recognition

Timothy J. Perfect, Nicola J. Weston, Ian Dennis, Amelia Snell
2008 Quarterly Journal of Experimental Psychology  
Here we replicate the two studies above, whilst manipulating the precedence (global or featural) of the letter stimuli in the orientation task.  ...  These data raise important questions as to what is transferred between the Navon orientation task and the face-processing tasks that follow.  ...  These data are not compatible with the assumption that a local or global processing orientation transfers from the Navon task to the face recognition tasks.  ... 
doi:10.1080/17470210802034678 pmid:18609403 fatcat:ju373dlt3vhibml2vqhp6j2nk4

Mitigating Gender Bias in Face Recognition using the von Mises-Fisher Mixture Model

Jean-Rémy Conti, Nathan Noiry, Stéphan Clemencon, Vincent Despiegel, Stéphane Gentric
2022 International Conference on Machine Learning  
In this work, we investigate the gender bias of deep Face Recognition networks.  ...  In order to measure this bias, we introduce two new metrics, BFAR and BFRR, that better reflect the inherent deployment needs of Face Recognition systems.  ...  this paper), which aims at deciding whether two face images correspond to the same identity or not.  ... 
dblp:conf/icml/ContiNCDG22 fatcat:emi5tmihljbbjgthtq3vmbjtve

Bias in Automated Speaker Recognition [article]

Wiebke Toussaint, Aaron Ding
2022 arXiv   pre-print
Despite their wide-scale deployment and known sources of bias in face recognition and natural language processing, bias in automated speaker recognition has not been studied systematically.  ...  Leveraging the insights from our findings, we make practical recommendations for mitigating bias in automated speaker recognition, and outline future research directions.  ...  Historical Bias Historical bias replicates biases, like stereotypes, that are present in the world as is or was.  ... 
arXiv:2201.09486v1 fatcat:ereutwo7vvb7bc7tlvwfsd5uxy

Bias in Automated Speaker Recognition

Wiebke Toussaint Hutiri, Aaron Yi Ding
2022 2022 ACM Conference on Fairness, Accountability, and Transparency  
Despite their wide-scale deployment and known sources of bias in related domains like face recognition and natural language processing, bias in automated speaker recognition has not been studied systematically  ...  Leveraging the insights from our findings, we make practical recommendations for mitigating bias in automated speaker recognition, and outline future research directions.  ...  Historical Bias Historical bias replicates biases, like stereotypes, that are present in the world as is or was.  ... 
doi:10.1145/3531146.3533089 fatcat:qwb2erisizgypbcn7i4kvwn7mq

Inter-racial contact and the own-race bias for face recognition in South Africa and England

Daniel B. Wright, Catherine E. Boyd, Colin G. Tredoux
2003 Applied Cognitive Psychology  
Own-race bias, where people are more accurate recognizing faces of people from their own race than other races, can lead to misidentification and, in some cases, innocent people being convicted.  ...  This bias was explored in South Africa and England, using Black and White participants.  ...  This does not mean that the bias always occurs, that it is large, or that it is equally strong for everyone.  ... 
doi:10.1002/acp.898 fatcat:yehoxlbp5jgploljpcpes6y56y
« Previous Showing results 1 — 15 out of 166,100 results