Getting immersed in a new subject is a great experience. ISO SC 37 wrapped up a week-long series of meetings yesterday in Okayama Japan. Here are some things I learned along the way:
SC 37 is smaller than the other ISO committees I’m in – about 60 in-person plus about 20 online this week across 6 work groups. It makes for a tight-knit group and provides lots of time to build relationships. It also means that the workload is sane – on other committees I have to ignore all but the 2 or 3 projects I’m directly involved in.
The main work of the committee is extensible data exchange formats, APIs, evaluation of biometric technologies, performance testing and societal aspects.
SC 37 was started in the early 2000s to meet the needs of ICAO to add biometric information to passports – face, finger and iris formats were the first. Since then, work has been done to cover other biometric modalities such as voice, skeletal structure, vein pattern and others.
Participants are primarily from Government border control, immigration, public safety, law enforcement departments. I’m one of a very small number focused on consumer-provided sensors and devices for use in commercial scenarios.
Biometrics is all about statistical analysis – I never really thought much about it – using the lazy term “probabilistic” – but it really is about estimation and weighting of data samples. So I might have to break open a textbook or two to relearn by sum-overs, means, medians, and standard deviations – not looking forward that that part!
“Convoluted Neural Networks” appear to be broadly used to look for patterns and help to draw conclusions. Another area I must learn about quickly.
There is a significant desire to apply biometric binding of people to data records – for example during ID Proofing and Verification. But it was clear that SC 37 is not the best place to do that work. SC 17 for mDoc credentials or SC 27/WG 5 for identity management seem better suited for this work.
I realized by the end of the week that being an “identity management” guy in the “biometrics” world is a very good thing – the latter can be applied to operations of the former and I get to work on both sides.
So I’m looking forward to working with the committee members and digging deep into the material. Hopefully I can bring useful opinions from the consumer world to drive discussions and improve the body of work.
IDIM Musings 