English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Current face recognition systems achieve high progress on several benchmarktests. Despite this progress, recent works showed that these systems arestrongly biased against demographic sub-groups. Consequently, an easilyintegrable solution is needed to reduce the discriminatory effect of thesebiased systems. Previous work mainly focused on learning less biased facerepresentations, which comes at the cost of a strongly degraded overallrecognition performance. In this work, we propose a novel unsupervised fairscore normalization approach that is specifically designed to reduce the effectof bias in face recognition and subsequently lead to a significant overallperformance boost. Our hypothesis is built on the notation of individualfairness by designing a normalization approach that leads to treating similarindividuals similarly. Experiments were conducted on three publicly availabledatasets captured under controlled and in-the-wild circumstances. Resultsdemonstrate that our solution reduces demographic biases, e.g. by up to 82.7%in the case when gender is considered. Moreover, it mitigates the bias moreconsistently than existing works. In contrast to previous works, our fairnormalization approach enhances the overall performance by up to 53.2% at falsematch rate of 0.001 and up to 82.9% at a false match rate of 0.00001.Additionally, it is easily integrable into existing recognition systems and notlimited to face biometrics.

Post-Comparison Mitigation of Demographic Bias in Face Recognition Using Fair Score Normalization