Communications of the ACM
In Facial Recognition Challenge, Top-Ranking Algorithms Show Bias Against Black Women
Even the best facial recognition algorithms still struggle to recognize Black faces, particularly for women.
Credit: Getty Images
Facial recognition algorithms have higher false-positive and lower false-negative rates for Black women, eyeglass wearers, and young children, according to the results of a facial recognition and analysis competition held during the European Conference on Computer Vision 2020.
The event called on participants to develop, test, and submit algorithmic methods with an aim of reducing bias in facial recognition technology.
Each team used the same dataset, containing 152,917 photos of 6,139 men and women ranging in age from under 34 to over 65.
When photos of different people were compared using the top 10 methods, women with dark complexions were discriminated against 45.5% of the time, compared with 12.6% for men with light skin tones.
Said the organizers, "Despite the high accuracy, none of the methods was free of bias."
From Venture Beat
View Full Article
Abstracts Copyright © 2020 SmithBucklin, Washington, DC, USA
No entries found