Old Story On Iomniscient At University Of San Francisco - Anyone Aware Of Long Term Success/Results?

There was a story posted on SecurityInformed regarding iOmniscient as a Facial Recognition solution installed at the University of San Francisco. Curious if anyone knows if this is still running and what the success rate has been? The following statements outline the use case and claim of creating a safer environment:

"Matching faces with an existing database, the system checks whether a person entering the building is authorized to access that facility. If an unauthorized person is detected, the system can track him or her across non-overlapping cameras. The information is also sent to CISCO’s Video Management System where it alerts the front desk attendant. USF implemented the system as part of its commitment to provide its students with a safer college environment."

"The system also increased the productivity of the staff. The University was able to achieve its objectives far more effectively and with fewer staff than if they had to perform this task manually."

"and USF is one of many happy customers who continue to implement advanced intelligent video solutions to ensure the safety of their students and the efficiency of their staff."

Very curious if the system was able to successfully verify each person entering a building as having authorization to access. If the system is checking against a "white list" of sorts, with all authorized persons enrolled, how would the campus security staff validate if a match percentage of a person entering is less than the acceptable match percentage (another words, they are not on the watch list)? Do they dispatch a guard to verify identity? How are students/facility and "authorized" individual enrolled, an integration to ACS still/2D flat photos or video enrollment? How would success be measured in a project such as this?

Would be interested in hearing of similar deployments and what worked / what did not?

Link to full article is below --

https://www.securityinformed.com/news/co-420-ga-co-1880-ga.15534.html

Login to read this IPVM discussion.
Why do I need to log in?
IPVM conducts unique testing and research funded by member's payments enabling us to offer the most independent, accurate and in-depth information.

Newest Discussions

Posts Latest
49
35 minutes by Undisclosed Integrator #12
1
less than a minute by Undisclosed Integrator #1
3
41 minutes by Michael Miller
1
less than a minute by Fredrik Ahlsen
16
4 minutes by Undisclosed #3