Debunking Start-Up's FST21 Case Study

By IPVM Team, Published Oct 07, 2013, 12:00am EDT (Info+)

This company won ISC West Best in Show 2 years ago, recently received $5 million in funding and is being talked about as the next big thing in security and surveillance for its 'revolutionary' incorporation of facial, voice and behavioral recognition.

Indeed, this summer, FST21 has been touting a large New York City apartment complex as a success story. Multiple security trade magazines have done case studies on the project so we wanted to see what all the excitement was about for ourselves. We sent a reporter to Knickerbocker Village to take video of the installation and talk to some of its tenants.

** **** *** *** ***** ** see ******** ** ******, *** ******** to ***, *** ******* *** **** different **** *** ******. ** ***** the ****** *** *** ** ******** or ********** *** *** ******* ** the **** ******* **** ** **** and **** ************, ***** ***** **** open, *** ******* *** ******** ***** were ***** ********* ********.

**********

************* ********* * **-******** ********* ******* **** houses  ~**** ******, **** *,***+ ****** and ***** *****. *** ***** ******** ****** is ******** ** ** ** **** way *** ********* ** ****** ***** apartment ******** ***** ****** *********** ******* of * ******* (***** **** *** system *** ** ********** *** ***** or ****** ***********). *** ******* ** simple: ** *** ****** **** * person ******** ** *************’* ******** **** the ****** ****** ****** *** **** and *** **** **.

******** *** ********* ** *** **** last ****, *** ** *** ********* by ********* *****. **** ****** *** replaced ** *** ********* ** **** year.

****** ***** / ***** *** *** Cards

***** *** ***** ** ****** *** system ** *** ** ********* ** the ******* (****** ** ** ******** at **** *** *********). ************* ***** to ***** *** *** ***** *** operate ****** ** ****** ***********.

****

****** *****, ***** **** ********* ** sales **** *** **** ** ************ this ********** ** *** **** (*****'* Digital ******* ********) ** $*,*** ****. 

“*** ***** ******* ** *** ************ yet, **** *** ***** ** ** stages, ******** **** *** ******** ******** entrances, *** **** ****** ** *** outside ***** ** ****,” ** **** of *************. 

*********

**** * ****** ***** ****** *** lobby, **** ** *** ******* ********. The *** ********** ***** *** ****** has ******** ********. 

 

*** ********* ***** ****** ******* **** it ******* * **** *** ***** when *** ****** ** ******* *****. Here ** *** ** ***** **** it *****:

***** **** ** **** ****** **** in ********, ******** *** ******** *** five ******* ****** ******* ****** **** queue **, ******* *** ***** ***** that **** ***** ****** ****** ** together (*** ******* *** *** ********) it *** *** ******** *** ** them. **** ***** ***** **** ****** opting ** *** ******** ******* ** facial *********** *** *** ****** ******* to ********* ****** **** ***** ****** walk **. 

*** ******* ** ***** ****** *** are ****** **** *****, *** ****** is ******** ** **** ** ***** to ******** ********. 

What ********* *** ** ***

******* ********** ********

*** ********’* ******* **** ******** ***** that ** *** *** ***** * ****** ********* ***** *** ******, *** ********* ** multiple *********, **** *** ****** *** initially ******** ** *** **********. "** the ********* * *** * *******, but * **** ** *** ****** and **** **** **** ** **," the ******** ***** ****. "* **** it. * ***** **'* ****," ** says. 

******* ******

***** ********* *** *** ********* **** problems ****** ** *** ** ***** the ****** *********** ******* *** **** to *** ***** *** *****. ****** have ********* ** ***** ***** ******** because ** ************ **** *** ******. In *** ***** *****, *** ******** says ** "****** ****" ************ *** based ** *** ********* ***********, ******* chose *** ** *** ******** *** facial *********** ** ***: 

FST21 ******** 

***** **** *** ******* ******** *** part ** *** ****** ******* *** the ******. “**** ** *** *** ****** *****. We **** **** ********** *****, *** for *** ***** *** ** ****** times *** *** **, *** ****** learns **** ******** -- ***** ** different ********* ******** ** *** ******** per **** ****** *** *** ** part ** *** ****** ********. **** the ****** ***** *** **, **** you ***** **** ****** *** **** we *** ***** *** ******** ***** of **.”

***** **** ** ***** ** **** to ********* *** ** ** *** working *** **** ********* ******* ***** able ** ******* **** ******** ********, but *** ***** ***** ***** ***** is ****** **** *** ****** ** enrolled. Peled **** *** ****** ****** **** a **.*% ******** ********* **** *** in *** **** ****** *** *** the ****** ***** ***** *** **** **** ***** *** enrolled. At ************* ****** *** ****** *** enrolled. 

**** *** ** ** ****** ****** said ** ****** ** **** *** facial *********** ******* ****** *** ****. For *** ******, *** ****** ****’* recognize ****, **** **** ***** ******** instead, *** **** ******* ** *** the ****** ** ***.

What *** ******** ******** ****

******** **** ** ** ***** **** how *** ****** *** ****** ** far. ** **** ** ***** **** the ****** ***** **** * **** accurate ****** ** *** ** ****** in *** *** ** *** ********. “We **** **** ********* ****** **** to ***** *** *** **** ***** cards ** ***** ****** *** **** we ***’* **** ****** ***’* ****** in ****** ** ****** *** **,” he ****. 

*** **** ****** ******* ** **** he **** ** **** ********* *** resistant ** *** **. **** ***** haven’t **** ** *** ***** ******** taken, he ****. “********* *’** *** **** people ** *** ********* *** ****’** say **** ****** ** ******* ** *****’* **** ** *’** say ***’* ** **** *** *** it. **** ** ** **** *** try ** ****’** ***** * ****** combative. ****’** *** ******* ** *** camera ** ******* ****. ****’** ******* a *** ** ********* ** ******* their ********. ****’** ****** ** **** it *** ****.”

**** ***** *** *** **** *** the ****** **** *****. “*****’* * ***** *** ***** **** *** ***** *** he **** *** *** ** *** waves ** *** ****** **** ** releases *** **** **** *****’* * security ***** ** *** ****** ******* the ****.”  

Camera *****

***** **** *** ****** ** ****** 75 ****** **** *** ***** *** can ******* ***** **** ********, ****** in *********** *** ****** ******* *******. Enrollment ******** *** ***** **** ** close ** *** **** ****** *** angle ** *** ****** ***** ** coming ******* *** ****, ** ****. As * ****** ***** ******* *** lobby ****, **** *** ** *** center ** *** ******'* ***** ** view (*** *****). 

*** ***** ******** ** ***** ** 12 **** **** *** ******, ** said. "** **** *** *** ****** recognition **** ** *** **** ******* path **** ****** **** *******. ***** ******** is *** ***, ******* ***** *** have **** ** ************* *******, *** know *** ******** ***** ** *** nearly ** ** **** ** *****, and *** ****** ***** ***** *****," he ****. 

***** **** *** ******* ****** ******** came **** *** *******'* *******, * former ******* *******. "****** ** **** all *** *** **** ***'* ***** to ****. *** *** *** ******** the **** ****," ** ****. "**'* a ********* ******* ** ********. *** technology ** ** ******** ******. ******** people *** **** ******* ********. **** we ***** ****** ** **** **** if *** **** ** *** **, show ******** ... **** ****** *** make *** **** ******* ****** ** be ********** **** ** **********." 

******

****** ***** ******** ********, * ****** of ****** *****:

*** *****

**** ** *** ****** ** **.*% accurate, ** ***** ******, *** ****** will ** *********** ** *** *****. FST21 **** *** ****** ** **** accurate ***** * ****** ***** ******* six ** ***** *****. ** **** people ******, *** ******** ** ***** to **** *****, ***** ** ***** to ** * ******** ********** ** new ********* *** *** ****** ****** with *** ******. 

********** ***'* ****

**** ***** ** ******* ***** **** at ** *** *****, ******* ***** face ** *** ******* *** ****** where *** ****** ** *******. ***** the ****** *** **** ******* **** properly ********** ***'* ****, *** ***** do **** ***** **** ********** ***** to ** ****** ********.

******** ***'* ********

***** *** ******** ******** ******* **** residents *** ****** ** ******* ***** faces **** ******* * ***, **** people ****** ** *** **** **** they **** ** ****** ***** ***. Others, *** *** **** ** ** so.

**** '*****'

******* ****** **** ** ******* ***** face *** ****** ***** ********, **** will **** * ******* ** ***** so (*.*., ********** ** **** *** their ******** *** ** ******* ******** ahead ******* ** ** *** ******). While **** *** ** ****** ** 'user *****', ****** ***, **** ** many *********, ** **** ****** * practical ****.

********* *** ********* *** *** ****

******* *** **** *** ****** ** supposed ** *******, **** ********* ** spoke ** ***** ********* *** *** card ****** *** *** *** **** to ****** ** *** ****** *********** system. **** **** ***** ******** **** the ******* ******* ** **** *** entrances ****** *********** ****. 

***********

** *** ****** ****, **********, ************ and ******* ***** ****** ** *** still * *******. **** **** **** people ******* ** *** ** ** as ** **** *****, *** ** one **** ****** *********** *** *** work *** * *** *** ******* a ******* *** **** **** ****** until ******* ******* *** *** **.

***** **** *** ****** *** ***** the ****** ** ****** ******* *** door *** **** ** ********* ******** ** ****** **** *** ****, but * ******* **** *** **** a **** **** *** ****** ** a **** ** ***** **** ****. 

******* ******* ******** 

*** ****** ***** ****** ** ****** that *** ** ***** *** *** not ********** *** **** ****** *** the ***** ** *** ******** ****** for ******. **** ***** ***** *** ******** we **** ****** ****** ** *** camera. *** *** *** ** *** right **** ** *** ****** * log ** *** ** *** ****** and ******** ********* ** *** ******** for **** ********.

* *********** ********** ** **** ****** is **** ** ****** ** * clear **** ** * ******'* ****** *********** ******* ******* ********. ** * person ***** ********** ** *** * hat ****** ***, *** ****** *** no *** ** ********* **** * person ******* *** ********. ** **** come ** ***** **** ******* ****, the ****** ***** ***** **** ******* piggybacked ***** *** **. 

"**** **'** **** ** ****** **** need * *** ***** ******** ** installed ******* ********* **** ****** *** number ** ****** ****** ******* *** door. ** ***** **** ******* *** number ** ****** ** *** *** if **** *****'* ***** ** ******** and *****," ***** ****. 

**********

** *** ******** ****, *** ****** works *** **** ********** ** ** outdoor ********. *** *********, ******, ** that ** ****** ******* ****** ***** to **** ********* ***** **** ** else *** ***** ** ********** *** inconveniences ******* ************. ***** *** ******** user '******', *** ********** ******, *** user **********, ***** **** * ****** for **** **** * ******** ****** in * ********* ***** **********. ** contrast, ** ******* **** **** ***** operate *** **** *********** ** * government ** ********* ******** ***** ***** would ** ****** *** ******* ** use *** ******.

Comments (4)

"“This is how our system works. We take your enrollment photo, and for the first six to eights times you use it, the system learns your behavior -- about 20 different biometric features we are learning per each person and use as part of our fusion platform. Once the system knows you it, lets you enter much faster and also we can raise the security level of it.”

Disregarding his ending that last sentence with a preposition...

It 'learns your behavior'? Is BRS Labs aware of this claim?

For a straight up enrolled-party matching database system, what exactly is it 'learning' that it doesn't 'learn' when the face is enrolled?

Check out their own "How SafeRise Works" video.... SSNs old reporter Dan Gelinas is in the video with the President of DCRM, the install co consulting with FST21 on the install in the vid. Notice the better camera placement that looks head-on at people walking in the entrance door.

Also, one of the big marketing pitches FST21 makes is having no need for cards or readers - so the Knickerbocker staff and residents must not have read the literature.

Also, note that in the How SafeRise Works video, they claim that if FR doesn't R, then the enrolled party can opt to use VR instead (which they show). Mute people, you're stuck outside.

Marty, thanks for sharing the video. The entrance in that scene is naturally more conducive to facial recognition. Because of the L shape to the entrance, when you first walk in, your head is directly facing the camera. Then you turn and go through the interior door. On the other hand, this camera is directly facing outdoors (when the door is opened) which increases the risk of WDR problems.

It's tough to navigate those issues - either angles are causing problems, or light is causing problems, or hats are causing problems, etc.

How acurate is it: If you only shave every four days. If you wearing a blond wig as opposed to one of the others. If you change to a different makeup with a reflective base. If if if if.

Ya can't even sneak into your own place (for watever reasons). Similar to:

"As a man pulls intothe driveway, his BMW reads a message to him from his wife stating her mother stopped by. The man gets seconds thought and decides to drive away. However, the next message asks, "Where the *honk* are you going?" The BMW Connected Drive reads email."

Or, After the Web Site access hiccups trying the the call trees, voice or key punch, and there isn't an option for your call. Just try to get to a human. And when you do are transferred three times having to repeat account numbers/reasaons over again. Only to find out the 24/7 system is having a problem. Or doing a 'routine' update. Best to call back later.

The more they try to make it easy. The harder it is....... 'to make it easy'. The question remains. Easier for whom.

More?

Did they mention NCIS showing their badge and guns drawn?

OK Just the fire department. OR,

You have 'Medic alert', and laying on the floor hepless waiting for an ER rescue team.

Oh wait! There was a bomb scare..... Next the residents will have to take off there shoes.

Can you picture one or more of the above in a SNL (Saturday Night Live) skit?

Hey. It's Tuesday am and I'm retired with a thousand other things todo. :-)

I just got a demo from these guys about 2 weeks ago and the system seems interesting, but I am not sure I can think of a use case for it where it would be more reliable than standard access control. It is also fairly expensive, but I suppose you might be able to offset that price by using the same camera for alarm call ups, but it does not seem to integrate with any current VMS or IDS systems.

Good luck to them, but I am not a customer.

Agree
Disagree
Informative
Unhelpful
Funny
Login to read this IPVM report.
Why do I need to log in?
IPVM conducts reporting, tutorials and software funded by subscriber's payments enabling us to offer the most independent, accurate and in-depth information.
Loading Related Reports