Avigilon Facial Recognition 'Appearance Alerts' Tested

By: Sean Patton, Published on Dec 18, 2019

Avigilon has released 'Appearance Alerts' in ACC 7.4, which adds real-time facial recognition.

To see how this performed, we tested them with our H4A and H5A cameras and ACC7.4, examining:

  • How well does Avigilon face recognition work?
  • Where does it fail?
  • What limitations exist for face watchlist management?
  • What distance/PPF is required for face recognition?
  • What is the maximum angle of incidence it recognizes faces?
  • How well does it work in low light/IR lighting?
  • How much does it cost?
  • How much additional resource on the Analytics Server is required?

Appearance Alerts adds to Appearance Search, which we tested here.

******** *********** '********** ******'** *** *.*, ***** adds ****-**** ****** ***********.

** *** *** **** performed, ** ****** **** with *** *** *** H5A ******* *** ****.*, examining:

  • *** **** **** ******** face *********** ****?
  • ***** **** ** ****?
  • **** *********** ***** *** face ********* **********?
  • **** ********/*** ** ******** for **** ***********?
  • **** ** *** ******* angle ** ********* ** recognizes *****?
  • *** **** **** ** work ** *** *****/** lighting?
  • *** **** **** ** cost?
  • *** **** ********** ******** on *** ********* ****** is ********?

********** ****** **** ************ ******, ***** ** tested ****.

[***************]

Search ** ******

***** ********** ****** ****** for ***** ******* *************** like *****, ***, ******, age, ******, *******, ***., Appearance ****** **** ****** on ******* *****. **** can ** ********* ***** they **** ******* ******.

Notice: ***** ****** **** ******

***** *** ******* **** strong, ** **** ** emphasize **** *** **** on * ***** **** of ~** *****, *** less **** *** ****** test ******** **** * short ****** ** ****. Given ******* ****** *********** issues, ** ********* ***** scale ***** ** ******* larger, ****** ***** ** verify *** ********* ******.

Detailed ********** ***** ********

*** ******* ********** ******* positives:

  • **** *********** ********** ** ~90PPF:** ***** **** **** recognition ****** ******** ** ~90PPF ** ****-*** ******. This ** ***** **** many ***** ******* **** require ****** ** ******, but ******* ** **** we ***** *************.
  • ****** **** ***** ***********:** *** *****, ******** recognizes ***** **** ****** confidence ** **** ****** and **** **** (******* from * - **' high, *° ** **° down ****, *** * - **°). **** ******* it ** **** ** recognize ***** ************ ** targets ** **°************* ** the ******, ***** ******* Anyvision (********* ****** *********** ******) *********** **** **** **** Recognition ****** (******)*****.
  • ***** ******** **** **% Mask ********:******** ********** ***** **% obscured ** * ****. This ** *** ***** face *********** **** ** found ****.
  • ******** ******** ** ********** with ****/**********:*** ****** ******* *********** (e.g. ***, **********), ******** recognized ******** ******* ********** or * *** ** high **** **** (~**°) but **** ***** ********** settings, ******* **** ** the **** **** ****** a **** ** *********** at ~**° ** ******.
  • *** ***** ********* ** the ******* ***** ** incidence, *** ***** ********:*** ***** *********** ** solid, ******* ** **** light *********** ** ***** systems. ** ********** ***** at *********** **** **** angles (~**°) ***** ******* toward *** ****** *** only ****** **********. ** recognized ***** ** ******* walking ** * **° angle ** ** **° down ****. ****** **** light ***********, ** ****** recognize ***** ***** **°*************.

*******, ***** *********** *********** was *****, **** **** management ** ******* *** has ******:

  • ** ******** ****** *********:***** *** **** ** added ** *** ***** list ***** ***** ******* by ********* ************ *******, not ******** ***** ***** (e.g. ********). ******** **** they ***** ** ****** this ********** ** *** 7.6, *******, **** ** a *********** ********** *** early ******** *** **** to ***** *** * person *** *** *** been ** *** ******** location ***.
  • ****** **** ********* ****:** *** *.* **** 1 ***** **** ** supported, ***** ******* ** all ***** ***** ****** as * ****** *****. This ******** ***** **** creating **** ******* *** specific ****** **** ******* multiple ****** ** **** person, *******, *** **** images *** ********* ** one ****. ******** **** they **** ** ****** support *** ******** ***** in *** *.*.
  • ********* ***** *** ****** Allowed:** **** **** ** create ********* ***** ******* and **** ********* ******. Avigilon **** **** ***** be ********** **** ** a ****** *******.
  • *** **** ****** ****** Supported:***** ******** ********** **-*** PPF ****** *** *** Face ***** **** ********* images, ** ***** **** the ****** ******* **** lower ********** ******. ************, reference ****** **** * significant **** **** **** contains ******* **** ****** (e.g. *****, ****, **** not *******) **** ********.
  • ****** *********** ***** **** Beards/Face *********:** ***** **** **** faces **** ****** *** any **** ********* ******** in *** ****** ********* the ***** ** * reference ***** ** *** Watch ****. ******** **** that ********** *** ***** of ********* *** ****** for ********* ** ***** but *** *** **** any ******** ******** **** beards, ****** **** **** and **** ****** ** visible.
  • ** ******** *********:** *** *****, ******* photos ** * ****** of ******** ** ****** displayed ** ****** ******* could ** **** ** trigger *********** **** ********** nearly ** **** ** the ******* ********** ******** the *****. ******** **** they *** *** ***** the ********* *** ******** detection ******* ** ** not ********* **** *** access *******.

Comparative **** *** ******

********** ****** ** ********* compared ** *** ******* number ** ***** ************ NVRs **** ******* ****** recognition *** **** *** relatively *********** ******** ** Western ******** ********* **** routinely ****** $*,***+ *** channel.

********** ****** ******** * per-channel **** *********** ******* which ** $*** **** with * ** ******* minimum.

Face ***** ********** ****** *********** *** **********

*** ********* *-****** ***** reviews **** ***** **** management *** *********** ***** from ******** *****:

*** ********* *-****** ***** reviews *** ********** ****** and **** *********** ***********, examining **** *** *** performance ******:

Face *********** ******** ***** **** ***** ***** *****

** **** *****, ******** performance *** *****, **** high ********** *********** ** subjects, ** *********** ****** of **********, **** ** this ******* ** ~**° perpendicular ** *** ******:

**** **** *** *******'* face **% ******* **** a ****, ******** *** able ** ***** **** a **** ***** ** confidence:

** ***** **** ******** was **** ** ***** subjects ******* *********** (*.*. sunglasses, ****, *****), *******, confidence ******* **** **** to ****** ** *** levels:

*** *********** *** ********** with ***** **** **** capture *** * *********** down **** ***** ** well:

**** ** * ******* result **** ** ***** in ************ ****, ***** **** ********* well **** * *********** head ****.

*******, **** ***** *********** was ***** ** *** light *** ** *********** scenes, **** * ****** decrease ** ********** **** high ** ******, *** no ****** ****** ** incorrect ***************:

Face ***** **** ************ *** ********** ******

***** *** ** ***** manually **** ******** *****, which *** *** ** the ******* *********** ** noted ** *********** ********** ****** ****. ********** ***** *** no ****** ** ******** select ***** **** *** analytics *** *** ************* detect:

*** ****** ******** ******* is ***************, *** ********* the ********* ** ****** we ***** ** *** previous ****:

*******, **** *** ** confusing *** ***** ** there ** ** *** to ******** ****** ***** faces ***/*** *** *****, there ** ** ********** of ******* * **** is ** ** *** detected, *** ***** ****** be ***** *** *****. This **** ********** ** ** ********** parameter ** *** *******'* appearance, ******* ** ******** color, ******/***** *****, ***.

************* ******** ***** *** also ** ***** ** the **** ***** **** from ********** ******, ** was ********** *******.

* ********* ***** **** could ******** *********** ***** positives, *** **** ***** list ******* ****** **** limited **** *******, **** significant **** **** ******:

************, ** ***** **** subjects **** ****** **** more ****** ** ** rejected, ************ ** ***** mouth *** *** ******* or ********, **** ** the ***** *** **** face ******* **** ***-******* subjects:

******** ***** **** *** ideal ********* ***** ****** have ~******, *** **** eyes *** **** *******, and **** ************ **** be ********* ** *** 7.8 **** ******** ******:

********* *** ********* ***** and ********** ******** **** are *** **********

  • ******* ** (******* **) pixels ****** *** **** ear-to-ear *** ****-**-*** ** the ********
  • ***** **** ****** ****** not ** ********, ****** include *********, **** ***** above *** *** ** the ****, *** ****** have **** *** **** visible
  • *** **********, ***** ** a ********** *** *******-**** score ****. ** ******** to **** **** *** weighting ** **** *****. For ****.*, ******* ***** are ***** **** ******** video *** ********* ***** criteria *** *******. ** ACC *.*, ******** ********* images **** ** ********* to * ****** ******* quality *** *** ** will ******* * ***** image ******* *****.

******* ********** **********, *** Face ***** **** ****** duplicate ***** *** ****** to ** ******* ** the ***** ****:

******** **** **** ** ACC *.* ******** ****** of *** **** **** and **** **** ** prohibited.

Duplicate ****** *** ** ****** **** ****** *******

**** *** ****** ***** List *** ********* ***** supported, ** ***** **** multiple ******* ***** ** alerted ** * ****** appearance ** * *******:

**** ***** ** ** issue ** * ****** subject *** *+ ********* images ** *** ***** List, ***** ***** **** the ****** ****** ****** list ** *** ***** of ********* ********* **** a ****** **********.

Face ***** **** ****** ******* / ** ****** *** *** **** ***********

**** ***** **** ***** filtering ** *** ******** ACC ***** ****** ***** well, ******* ********* *** matched *****:

*** ******* ********** **** is *** ******* *** not ** ******** *** a ****** **** *****, so * ****** **** 1000s ** ******** *** day/hour **** ** ****-********* to ****** *******.

************, ***** ** ** function ** *** **** Watch **** ********** ****** to ******* *** *********** of * *******. ******** said **** **** ** added ** *** *.*:

**** ******** ** * real-time ********** ******** ** this ***** ** ****. In *** *.* *** will ** **** ** launch ********** ******** **** the **** ********* *** also ****** ****** ** launch ********** ********.

Comments (19)

***** **** ** ****** the *** *.* *** don’t **** ** ***** up ******* ******.

**** ** *******, *** need ** **** ********** Search ******* **** ****** GPU, *** ******** *** Face *********** ********.

****** **********. ***/***, ****** with *** *** **** Rec *******. **** *** is *** ** ****** yet ** *** ** Appliance.

****** ****!

*** ***** ** **** those ******* ***** ** his **** :)

* ***** ***: *** system ******** ****** ************ well ** ******** ***** a ***** ***** ***** be **** ** ********* the ****** ** *** picture. * ** ******* about *** *** ***** picture ** *** ***** where ***** *** ********* no ****** ******** *******. Considering *** (************) *** reference ******** ** ***** it **** **** *********.

**'* **** **** *** bad - ******** *** low-light *********, *** ****** that ** **** ******* with * **** **** here. ** *** ****** says "**** ** ***", but *** ***'* ********** present * *****/******** **** a ***** ***** ***** presented *** (**'* **** a ****** *** ****) - *** *** *** it ** *****?

* *** **** **** it ********** ********** ****, so *** *** ****** ever****** ********* ******, ** does ** ****** ******* the **** ******** ***** when ******* ***** **? E.g. *** *** ** people ** *** ********* database, *** ** **** were ***. **** *** of *** ** ***** by, **** ** *** "Oh, ****'* ******* ** don't ****" *** ***** that **** ****** (**********) if *** ** *** 30 ***** ** ** low *****, ******** ******, new ****-*** ***. ********* how ***** **** ** miss * *********?

** ***** *** ******** option? ** *** ****** says "****'* ***" *** in ******* **'* ****. Can *** **** **** the ****** "**! *** system. **** ** *** Rob!"?

* ***** **** *********** was **** ******, ********** considering (** *** *****) how *** **** ** the ********* ****** ****, on *******. *** *********** limitation (***** ** ***** Avigilon **** **** ***** be ****** ** *** of *** **** *******) is ********* ****** *** limited ** **** ***** capture ** *** ********* cameras, ***** *** ********** impact ********* **** ******* reference ******.

* ******* *** **** around ***** ** ** print *** ** ***** of *** *******, **** it ** ** ***** of *** ** **** cameras *** ******** ***** the **** **** **** image. ** *** *** test ****, *** ******** does *** ********* ****, I **** *** ** I *** *** **** to **** *****.

********* "******" ***********, ** a ****** ******* * person, *** **** *** not ******* ** * face ** *** **** Watch ****, *** **** the ****** ******** ***** is *********. ** * person ** ******** *** matched ** * **** in *** ***** ****, only *** **** ***** list ***** ** *********. There ** ** "*******" or "*********" **** *****.

******* ** *** ***** it ****** * *********, that ******* ** * couple *******. *** ******* match ********** *****, *** the ************* ** *** H5/H4 ****** ********* (****** ****), ***** ******* **** well. ** *** **** to ******* * **** or ****** ********** *****, to ***** **** ******** false *********, ***** ** our ******* *** ***** not ******* * **** match ***** *** ******* wearing * ****. *******, the ****** ** ***** detecting * ****** ** those **********, ** * "Person ********" ***** ** still ********* ** *** system. **** ** *** case *** *** **** capture ***** *** ******* confidence ***** ** *** met.

***** ** *** *** training *** ******** *****.

* *** ** * reference ***** ** *** holding *** ******* ***** as ** *** ** the ******* **** ** capture * *****, *****-****** image ** *** ** the ******. **** ** not * *********** ******** work-around, * **** ****** to *** *** *** system ***** ****** **.

** ******, ******** *** at * ****** ********** level:

*******, ** ******* ** a **** ********** ***** with * "****" ******** image:

*** *** ** ***** in *** ******* ********* image, *** ***** *** the ******* ** *** had ** ***** * reference ***** ** *** Watch **** *** ******* that *** *** **** captured ** **** *******.

*** *********** **** ***** now ** ** *** roadmap **** *.*? ** have ****** ** ********* as ** *** **, but ** *** ******** Avigilon *** ** *** same ***** ** ***** be ****** *****.

**** *** *** ******* to ** **** *** integration **** *****? ****, any ******** ** ******* Lenel **** *** ** get *** *** *********** your ******* ***.

********* ***** **** **** as **** **********. "**** at *** ****** ** unlock *** ****" **** of ****.

***** ** *** ****** standard ** **** ******** (at **** *****) **** looking ** ******** ****.

**** **** **** ** your ***** **** *** to ****** * **** that *** ******** *** use *** ******** *********. Other **** ****** *** grab *** **** **** your ******** ******* *** load *** *** ** an **** *** *** your ***** ****...

* ******* ********* *** that ******* ** * recall *********, *** **** point *** *******! *********, Avigilon ******* ** *** as ****.

*** **** ****** ******** only ******.

* *** **** *** my ****** **** * photo ** **** ***** turnstile ** ***.

* ******* ******** ***** this *** **** **** their ********.

******** *********, **** *** summary ** ***** ***********:

Video ******* **** ***** *******

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

  • **** ** * *** integration ******* ******** ******* Center * (***) *** later, *** ***** ******* 7.2, *.*, *.* *** 7.5.
  • ** ********* *** ******** functionality, ******** ****** **** the ***** ******* ****** to ** ********* *** recorded ******* *** ******, as **** ** *** ability ** *********** ******* individual ****** ** ***** control ****** **** ** the ***-****.

Video *********** **** ***** *******

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

  • **** ** * *** integration ******* *** *** system *** ***** ******* 7.2, *.*, *.* *** 7.5.
  • ** ********* *** ************* provided ** ******** ************, allowing ***** ** ****** ACC ******** ****** *** OnGuard *********. ***** *** view **** *** ******** video *** ******, ****** video, *** ******* *** devices *** ******* ********* events ********* ******** ***** through *** ******* ********.

*** ** **** ** you **** ***** ******** integration ******* *********.

****, ****** *** ******* that ** ****! **** replied ** ******* ***** that **** * ** really ***** ** ** there **** ** *********** with ***** **** ****** Avigilon **** *********** ** unlock ***** ****** *****.

******* ****** *** **** that's * ***** ******* is *********:********* ********* *** - Lenel.com

** **** **** **** with ***** ***** ******* or **** ******** *******?

Login to read this IPVM report.
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.

Related Reports

Facial Recognition Systems Fail Simple Liveness Detection Test on May 17, 2019
Facial recognition is being widely promoted as a solution to physical access control but we were able to simply spoof 3 systems because they had no...
Carnegie Mellon AI Startup Zensors Profile on Jun 11, 2019
Zensors is a startup formed by Carnegie Mellon graduates from a Carnegie Mellon research project, offering customized models per camera that they...
Anyvision Facial Recognition Tested on Aug 21, 2019
Anyvision is aiming for $1 billion in revenue by 2022, backed by $74 million in funding. But does their performance live up to the hype they have...
Verkada People And Face Analytics Tested on Aug 16, 2019
This week, Verkada released "People Analytics", including face analytics that they describe is a "game-changing feature" that "pushes the...
Scylla AI Video Analytics Company Profile on Aug 29, 2019
Scylla, an AI analytics startup, says they are targeting 1 Billion dollar valuation in 5 years and it "is not rocket science" to detect weapons and...
Hikvision Dual Lens Face Recognition Camera Tested on Nov 19, 2019
Hikvision's Dual Lens Facial Recognition camera, claims that it "adopts advanced deep learning algorithm and powerful GPU to realize instant face...
Avigilon Appearance Search Tested on Oct 30, 2019
Avigilon Appearance Search claims that it "sorts through hours of video with ease, to quickly locate a specific person or vehicle of interest...
Rhombus Cameras, VMS and Analytics Tested on Nov 06, 2019
Rhombus boasts they have created "the new standard in Enterprise, cloud-managed video security" and told IPVM in January 2019 they offer twice the...
BriefCam Video Analytics Tested on Jan 06, 2020
BriefCam, acquired by Canon in 2018, is one of the most commonly used video analytics offerings in the West. But how well does it work? We...
Bosch / Milestone Forensic Search Tested on Jan 08, 2020
Bosch's Forensic Search Milestone plugin integrates Bosch IVA and EVA analytics search in Milestone XProtect, claiming to "gain complete control on...

Most Recent Industry Reports

Masks Cause Major Facial Recognition Problems on Feb 24, 2020
Coronavirus is spurring an increase in the use of medical masks, which new IPVM test results show cause major problems for facial recognition...
Every VMS Will Become a VSaaS on Feb 21, 2020
VMS is ending. Soon every VMS will be a VSaaS. Competitive dynamics will be redrawn. What does this mean? VMS Historically...
Video Surveillance 101 Course - Last Chance on Feb 20, 2020
This is the last chance to join IPVM's first Video Surveillance 101 course, designed to help those new to the industry to quickly understand the...
Vulnerability Directory For Access Credentials on Feb 20, 2020
Knowing which access credentials are insecure can be difficult to see, especially because most look and feel the same. Even insecure 125 kHz...
AI/Smart Camera Tutorial on Feb 20, 2020
Cameras with video analytics, sometimes called 'Smart' camera or 'AI' cameras, etc. are one of the most promising growth areas of video...
China Manufacturer Suffers Coronavirus Scare on Feb 20, 2020
Uniview suffered a significant health scare last week after one of its employees reported a fever and initially tested positive for coronavirus....
Cheap Camera Problems at Night on Feb 19, 2020
Cheap cameras generally have problems at night, despite the common perception that integrated IR makes cameras mostly the same, according to new...
Milestone Launches Multiple Cloud Solutions on Feb 18, 2020
Milestone is going to the cloud, becoming one of the last prominent VMSes to do so. Milestone is clearly late but how competitive do these new...
Video Surveillance Architecture 101 on Feb 18, 2020
Video surveillance can be designed and deployed in a number of ways. This 101 examines the most common options and architectures used in...
UK Stands Behind Hikvision But Controversy Continues on Feb 18, 2020
Hikvision is exhibiting at a UK government conference for law enforcement, provoking controversy from the press, politicians, and activists due to...