Biometrics

WPF executive director Pam Dixon to present lecture at National Academy of Sciences on the future of data science and privacy

Executive Director Pam Dixon to present a featured lecture at National Academy of Sciences, Engineering and Medicine at its Colloquium: The Future of Data Science  Pam Dixon’s lecture will be Wednesday October 28 from 12:00-12:45 am Eastern. The lecture focuses on forthcoming and converging trends, and is titled Looking Ten Years Ahead: Key Converging Technologies in Computing, Data, and

WPF advises that DHS biometric collection proposal needs scientific, factual, ethical, and human rights basis, urges creation of transparent multistakeholder process to work on ethical guidelines regarding childrens’ biometrics

The World Privacy Forum filed comments with DHS regarding its proposed new rules regarding biometric collection, and asked DHS to reconsider the rule and to apply the science, ethics, and international conventions regarding the protection of victims of human trafficking to the rule prior to moving forward. WPF found that the proposed rule avoided discussion of

WPF to discuss COVID-19 lessons learned and challenges ahead at OECD – Global Privacy Assembly Workshop

WPF Executive Director Pam Dixon will be presenting at an OECD and Global Privacy Assembly workshop on the risks, challenges, and potential solutions regarding the intersection of COVID-19 and the uses of identity in a global public health crisis. Event details: COVID-19 and privacy virtual Workshop on “The road to recovery: Lessons learned and challenges ahead”, hosted

Face Recognition and Face Masks:  Accuracy of face recognition plummets when applied to mask-wearers; NIST report 

NIST has published its first report regarding face recognition algorithms and the wearing of face masks. The report quantifies how one-to-one face recognition systems perform when they are utilized on images of diverse people wearing a variety of mask types and colors. The study found that pre-COVID-19 FR algorithms have substantial error rates, some reaching as high as 50 percent for false non-match rates.