Demystify Biometrics introduces toolset to analyze, compare algorithms

A new resource has been launched by Demystify Biometrics to apply its mission of increasing transparency to governments, enterprises, analysts and investors around biometric performance to the benchmarks created by organizations like NIST.

The new Biometrics Intelligence Hub now presents overviews and analysis of NIST evaluation reports on algorithms for 1:N face biometrics.

It will soon add the results of biometrics evaluations for 1:1 verification, liveness detection, age estimation and other modalities, Demystify Biometrics Founder Ashok Singal said in a LinkedIn post. Singal expanded on the plan in an exchange of messages with Biometric Update, noting that evaluations of face morphing detection, demographic performance differentials and information about image quality standards are also on the roadmap. Eventually, he says, the Hub will include field relevance indicators and publicly verifiable reference deployments.

The idea is in part to help buyers “ask better questions,” Singal told Biometric Update. “Not just ‘Who is #1?,’ but ‘Who is reliable, scalable, compliant, and suitable for my use case?’” The Hub does not rank or recommend biometrics vendors, instead offering transparent analytics to help organizations make informed choices.

Singal believes the industry needs to further demystify nuances like how algorithm performance evolves over time.

To that end, the tool provides news report alerts detailed developer profiles and performance reports, as well as a function to search for algorithms based on performance, geographic or use-case criteria. It enables competitor benchmarking and allows organizations to compare biometric algorithm performance across different datasets.

Article Topics

biometrics  |  Biometrics Intelligence Hub  |  biometrics research  |  Demystify Biometrics  |  face biometrics  |  facial recognition  |  NIST