Agenda:

https://github.com/biometricITC/cPP-biometrics/wiki/iTC-Meeting-Agenda-2020-11-19

Call started at 10:00 am EST

Attendees

• Brian Wood
• Jonathon Rolf
• Naruki Kai
• Clare Olin
• Fiona Pattinson
• Greg Ott

Record of Decisions

• None on this call

Action Items

• Brian will create a pull request for a new table set in the fingerprint toolbox

Minutes

The call started with a quick review of the Task List. Brian had updated the eye, face and vein toolboxes to show as completed.

Brian noted that the presentations were made at the CCUF and ICCC over the last week and that they seemed to go well. There weren't many questions. The CCDB did ask about the plans on proceeding with CMFA, but it was more about process at this point (timeline, etc).

Brian did mention that he had a question about a possible need for use case 2, though based on further discussions it may be more about application biometric testing. As the project is currently on hold, it isn't clear if/when this will be raised for further review.

The call then moved on to the fingerprint toolbox review. The first topic was a quick review of 3 pull requests that had been merged (#24, #25, & #26).

The call then moved on to Issues #16 & #17 to discuss how to best represent the attack potential tables for fingerprint. The mold and cast are much more separate than in the other modalities, and so presented as separate calculations. It was decided in the end that possibly the best answer would be to create 3 separate tables. One would be the calculated attack potential, which would be based on taking the highest score from each component for each factor. The other two tables would show the scores specifically for each component. Brian said he would create this after the call for one of the attacks so everyone could review an example before it was spread to the rest of the attacks.

The group then discussed Issue #18 about the proper score for getting the fingerprint. The discussion covered whether it should be assumed to focus on latent prints or others (such as high quality photos directly taken or gathered from public sources). After some discussion it was felt that given the tests are using cooperative molds, meeting the non-cooperative gathering of data would still be within the parameters of what is being tested, and the current scoring would still be within the basic attack potential. It will be updated once the tables have been finalized.

The final topic of the call was Issue #6. The agreement was to look into adding some additional text in the fingerprint overview about possibly using other materials than those specified. Further, Brian suggested the iTC create a toolbox update process (sort of like the IT, but specifically for incoming changes for the toolboxes). This would lay out the expectations upon anyone submitting the changes (specifically on what needs to be provided for a timely response and how it can be claimed in an evaluation). Brian will look into creating this document.

At the end of the call a request was made for everyone to review Issue #5 to see what may need to be added here (or to the toolboxes in general). This will be the first topic of the next call.

The next call will be on December 3.

The call ended at 11:02am EST.