Axxon Face Search Tested

Author: Rob Kilpatrick, Published on Oct 26, 2018

AxxonSoft has brought facial recognition to their Axxon Next VMS for free with the simply named Face Search, claiming to allow users to find subjects across multiple cameras to drastically cut search time.

ipvm tested 2

In this report, we provide our full test results on Axxon's Face Search, examining:

  • How many false negatives or positives to expect?
  • How well does the analytic handle subjects looking down or away from the camera?
  • How does recognition perform in low light, <1 lux with IR, and WDR scenes?
  • How does Face Detection impact CPU usage?

Readers interested in facial recognition should also see our recent test of Dahua's Face Recognition Camera, the HF8242FN-FR.

************ ******* ****** *********** ** ********** **** ****** **** **** *** ****** ********* ******, ******** ** ***** ***** ** **** ******** ****** ******** cameras ** *********** *** ****** ****.

ipvm tested 2

** **** ******, ** ******* *** **** **** ******* ** Axxon's **** ******, *********:

  • *** **** ***** ********* ** ********* ** ******?
  • *** **** **** *** ******** ****** ******** ******* **** ** away **** *** ******?
  • *** **** *********** ******* ** *** *****, IR, *** *** ******?
  • *** **** **** ********* ****** *** *****?

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

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

Summary: ******** ****** ***** ******* ** ****** ********

***** ** *** *****, ***** ****'* **** ****** ****** ****** well ** ******* **% ********** ********, **** **** ******** ******** matched, **** *** ****** **** ********** **** * **** ** testing ** * ***** ****** ** ~**-** ******. ***** **** Search *********** **** **** **** ******** ** **** ******** ******** across ******** *******.

************, ********* ********** ****** ** ***** ***** *** *** ********** than ***** **** *** ************ ** **** ***** (*** *** test *******'* **** *** ******), **** ***** ***** ****** ********** ** * scene.

Some ******* **** *** *********

*******, ***** *** ******* *** ****** ** ** ***** **:

  • **** ***************:**** ******** **** ********** ******** *** ****** ** *****'* **** search, ** **** (>**%) ********** *******. **** **** ***** ******** with ******* *** ** *** *****. ***** ***** ********** ********** (~60-70%) ******** **** **********, ** ****, *** ***** ****** *** below *****'* ******* (**%).
  • ******* ***** ****** **********:** *** *****, **** ***** **** *** *********, **** ****** similarity ********* ** ~**% ** **** **** ******** ****** **** or **** **** *** ******, ************ ******* ******* ****** ******** (~5-10% ** *** ****).
  • ***/*** ***** ********** **********:************, ********** *** ******* ** ~**-**% ** *** ***** (~*.* lux - * ***), ***, ** **** ***** ********** **, similar ** ***** ********* ** **** *****.

Upcoming **** **** *** *********

***** ***** ****'* ****** *********** ** ******* ** ******** *** in ******* ********, ********* *** ***** ** ******* **** ********* in *** **** *** ******. **** ****** ** ***** ******* of ***** ********* *** ***** *********** ****** ******* ** **** Search, *** ** ** ******* **** *******/************ **** ** ** final ********** ********. ** **** ** ****** **** ******** *********.

*******/************

**** ****** ** ******** ** ** ********** **** **** ***** Next ******** ******** (~$*** ****/*******) *** ** ** ***-** ******* when ***** ***** **** ************. ***** **** ***** **** *** support **** ******. ****** **** **** *********** ********* **** ******* additional *********.

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

**** ****** ** ****** ** ***, **** ***** ********* *** or **** *******, ********* * ********* ***** ** *** *******, and ********* ********** ********* (** *********). ********** ******** ** **%, which ** ***** ********** ******** ** *** *****, **** *** mistaken ************ ** ******** ***** **** *-* ********, ********* *****.

*** ***** ***** ************ *** ** *** *** ***** **** Search *******.

Configuration ********

**** ****** ******** **** ********* ** ** ******* *** **** camera, ***** *********** **** ******** *** **** ** ** *****, analysis ********** *** **** *****/****** ******** ** *********, *** ********* enabled.

**** **** ******** ********** *** ******* ****** *** ***** ** cameras ***** *****. *-*** ****** ** *** ***** ****** *********** CPU ******, ********* ******* ***** ** **** ******.

*** ********* ***** ***** *** ******* ************* ** **** ********* on ** ********** ******.

Solid ****** *********** ***** ******* ** ****** **********

***** *****'* ******* **% ********** ** ******, **** ******** **** properly **********. *** *******, *** ******* ***** *** ************ ********** with >**% ********** ****** ******** ******* ** ******** ********** *** at ******** ******, **** ** ***************** ** ***** ********.

most subjects recognized with high similarity

Bearded *** *************

*******, ******* ******** **** ********** ********** ** **** ***** ** a **** ********** *****. *** *******, *** ******* ***** ** recognized *** **** ** ******* ** **% **********, *** ** another ******* *** ** *** **** *****. ******* ******* **** seen **** ***** ******* *** ****** *******. **** **** ** other ****** ******** **** ** ******* ** **** ****** ***************** in *** *****.

bearded subject

Lower ********** **** ***** ******

*** ******* ****** ** **** ****** ********** *** **** ****. For *******, **** ******** ****** **** ** ** *** **** (as **** ** **** ******* ********), ********** ******* ** ~**-**% or ****. *******, ****** *** ****** **** ** *****'* **** Rec ******, **** ****** ****** ******** ****** ****** **********, ********* <10 ***** ** * **** ** *******.

head titled 2

<1 *** ********* ********, *** ******* ******* **********

***** ******** *** ********** ***** ** ~*.** ***, ******** ********** was ******* ** ~**-**%. **** ** **** ** ******** ** the ***** ******, ***** ****** ** ******/********* ******** ** lux.

low light 2

CPU **** ******

*****'* **** ********* ********* ********* **** ** ~*-*% *** ***** stream ***** ******** ** *** *****, ***** ** ***** ** 5557U *********. **** ******** ** ~**% ***** *** ***** *** four ***** *******, * *********** ******** **** ******* *** ******* Face *********, ***** ******** **** *-*% ****.

*******, *** ***** ****** ******* **** ******* ***** ***** **** analyzed. ****** *** **** ***** ** **** *******, * *** 5 **, ********* *** ***** ** **-**%****, ********* ** **%+ *** ***** *****. ***** ***** ****** resolution ******* ** ***** **** **** ****** ***** ****** ****** of ***** *********. ***** ********** ******* *** ** ********, ****** with ********* ***** *************** *********.

CPU ****/*** ******* *******

*********, ***** ********* **** ********* *** *********** **** ** *** server's ***. *******, **** ***** **** **** **** ** ********* Intel's ******** ******** ****** ******* ** *** **** ****** ** increase **********, **** ***** ******* *** ****.

Wide ***** ** **** / *** *** *** **********

**** **** ***** **** ****** *** ** *** ** *** camera **** **** ********* *******, ***** **** **** ****** ** view ****** *** *** ******* *** ********* ******* ** *** tests, *.*., ** *** ***** ** **** *****.

wide field of view

***** ********* ************ ** ** ****** ******* *** **** *** detection, ** ******* ***-*** ***. ************, **** ** *** ****** are **** ****** *** ********, ***** ****** ****** ** *** CPU ****** ** ***** ***** *******, ** ***** ****** ****** levels *** ***** *** ***** ******.

Versions ****

*** ********* ******** **** **** ****** *******:

  • ***** ****: *.*.*.*.****

Comments (13)

Undisclosed Manufacturer #1

10/26/18 07:03pm

"**** ****** ** ******** ** ** ********** **** **** ***** Next ********"

*** *** **** ***** ****? *********** * **** ** **** it ** ******** ** ***** ******** ******* (**** ********* *******), is ******** ** ***** ************ (** ** ***** **** ****** this ********** ** **** ** ********) *** *** ********* ** the ***** ********* ***** *******.

** ** ********* *** "****" ** *** **** ** ** to *** **** ********* ******* ** **** ** *** ******* feature *** ****. * ***** ** ***** ** "****" ** you ******* ****** *** ******** ******* *** **** ***** ******** and *** **** ******** *************.

*** *****://***.*********.***/********/**********/********.***

Avatar

Alan Ataev

In reply to Undisclosed Manufacturer #1 | 10/29/18 06:39am

**** ****** ******* ********** ** **** ** ** **** ***** was ******** **** **** *** *** ******** ******** **** ******* tracking **** **************, ******** ****** ** ******, ****** **** **** human/vehicle, ***** *** ***. **** ****** ** * **** ** our ** **** ** **** ** ******** *** ******** ** Professional *******.

Sergey Bystrov

NetworkOptix | 10/26/18 03:24pm

***** ****, **** *** ***** *******!

Avatar

Murat Altuev

AxxonSoft | In reply to Sergey Bystrov | 10/29/18 05:58am

*-*% *** ***** ** *** ***, ********, *********** **** ** need **** ** ********** ******.

**** ** *** ******** ********** ** ****** ******** ********* ****** of ******** *** ********* **: *****://****.*********.***/****/**********.***

*** ****, **** *** ** ******* ** **** ***** *** need ******** ******.

**, *** *********** ** ********* * ***** ***** ** *** this ********** *** *** *** *******, ***** ** **** ** available **** ******* **** *** ******* ** ** **** **** all *** *********, **** ****** *******, *** *** ** **.

*** *** ****, **** ***** *** **** ************ (******** *** hardware) ****** **** **** ** ******* *** *********** ************.

Sergey Bystrov

NetworkOptix | In reply to Murat Altuev | 10/29/18 08:06am

** **** **** ** ********** ******.

***** **** * ******, *** **** *****.

Avatar

Murat Altuev

AxxonSoft | In reply to Sergey Bystrov | 10/29/18 10:13am

*** *** ** **, *** ******* ***** ************* ** **** case. *** **** *********** ** **** **** **** ********, ** minimal *********** ** *-** ***.

Avatar

Richie Perna

10/27/18 01:56am

** *** ****** ***** ************** *** ***** ** *** **, we **** *** ******** ******** *** ******** **** *** ******* on **** ***** **********. *** **** ********* *** *********** ********** is ****** ** ****. ** **** ******** ******** **** ********** in ****** ******* ******** ************ *** *** ********* ********* ******** from *** ******** ****** *** *******. *** ********** ** ********* being **** ** ***** *** ************ ** **** ** **** large ***** ***********.

** ** ******** ** ******* *** ******* ******** ** ******* this ******** **********. **** ***** ********* ** **** ***** ********** accuracy. ** *** **** *** **** **** *** **** **** Search ******* ********* *** / ****** ************** *** ** *** having ********** ******* *** *** ******* ******* ** ********** **** Axxon ** ******* **** ******* ** ******** *********.

***** ****** ********* ** **** ******* *** * ***** ********* of *** *** ********** ***** *** *** ******* ** *** it **. ***** ** *** **** **** ** * ***** review **** ******** **** *******.

Avatar

George Skouras

10/27/18 08:04pm

******'* ***** **** **** **. *****.

***** *** ************ ** ********* ** ****** *** ******, ** have *** *** ****** ** ****** **** ** *** *********** partners/resellers ** *******-** **** ******** ******* -***** ******- **** ***** success *** ************ ****** **** *** ***-********* *** *** ******** managers *** **** **. ***** *********, ******** *********** ** ********* and ***** *****-***** ************** **** *****, ******** *********, **** *****, military **********, ** **** **** ********.

***** ** *** ** *** *** ***** ******* (***** *** IP ******* *** ********* **** ***** ********* *.*******@******) ** ******** almost *** ******* **********. **** *** ********* ***** ** *** maximum ******** (*** *** ***** ** ** **********) *** *** minimum ********** ******** ** ** ****, ** ********* *** *********** hardware *** *** ******* (********* ******** * **** ***** **** to ** **: *****://****.*********.***/****/**********.***). *** ******* *** *******.

**** *
*************** ** *** **** **** *** *** **** **** ******** test. ********* ****, **** ****.

**** *
********* *** ******* *** *****************, ** *** ****'* ***** **** these *** *** *** *** ****** *** **** ****** (******** at * ********* ***** ** ****, **** * ****** *****, from * ********* ***** *** ****** ****** *** ******** ******** etc.), * ******'* ** ********** **** ****** **** *** ***** belong ** * ********* ***. ** ** *******, ** **** close **** **********, ** * **** * ******** ******* ********* for * *******, * ***** ****** **** ** **** ** opinion ** ** *** *** *** *** *** ********* ******.

@*********** ************ #*
**** ******* **indeed ******** ** ***** **** ************ ******* as ** ***** **** *** ****** ****** ("Extra Feature Package"). With this feature you can search for a face in a camera. And indeed can't be included in the Start Version.
*** ***, **** ******* ** ** ******* ******** ** ***** Next ******** ******* **** ** ********** ****,but **** ** * "*****-****** **** ******", which means that you can search for a face in many (** ***) ******* ************** and also have the exact spots (where the faces are recognized/found) pinpointed on a 3D map. This feature is extremely helpful for the operators.

Undisclosed #2

10/29/18 08:15am

***** * ****** *** *** * **** ***** ** *** reference ****, *** **** ****** ** **********, *** ** ********** at ***, ***** **-***** *** ***?

Avatar

Murat Altuev

AxxonSoft | In reply to Undisclosed #2 | 10/29/18 08:16pm

********, ********** **** ** ***** **%, *** **********, **** ******, he **** ** *****.

Undisclosed Manufacturer #3

10/29/18 09:46pm

* **** ** **** *** * ************ **** ****** * little ***** ***** *** ** ***** **** ******* **** ******* every ****** ** **. ** *** **** ****** ** ******* people's ********* *** **** **** **** ** ** *** ****, but *** ** ****** ** **** **** ** *** **********.

Igor Falomkin

AxxonSoft | 10/30/18 06:00am

**** ****, ***** *** *** **** ******** ****.

*** ********* **** ********** ******* *** **** **** * ***** was * *** **** *** **. ***** *** *** ****. I ***** **** *** ****** ** **** ** ** *** have ****** ****** ** *** **** * ***** ** *** training *** *** **** *********** *********. ** *** ***** **** learning ********* ** *** ** ***** ****** - ** *** going ** *** **** *********** ****** **** *** ******** ***.

** ******** *** *********** ** *** ***** ** *** ***** OpenVINO ******* **** ******** ****** ********* ************** ** **** ******** inference ******. * **** ** **** *** ** ** ******* more ******** ** *** **** ******.

Avatar

Nabil Reigber

AxxonSoft | 10/30/18 09:26am

****** ** *** **** ****, **** ********* *******! *** ****** free ******* ** ***** **** *** **** ********, ***** ******** all *** *** ******** ******, *** ** ********** **** *** AxxonSoft *******. ** **** ** ****** ** ***** **** ******* that ***** **** ** * ****** ****** *********** *** ********* we **** ****** *** ******** ******** **** *** *********** ********** where ** **** *** ********* *************** *** **** ******* *** clients. **** ** *** **** ** *** ******** **********:

*****://****.*********.***/****/**********.***

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

The IP Camera Lock-In Trend: Meraki and Verkada on Jan 18, 2019
Open systems and interoperability have not only been big buzzwords over the past decade, but they have also become core features of video...
NYPD Refutes False SCMP Hikvision Story on Jan 18, 2019
The NYPD has refuted the SCMP Hikvision story, the Voice of America has reported. On January 11, 2018, the SCMP alleged that the NYPD was using...
Mobile Surveillance Trailers Guide on Jan 17, 2019
Putting cameras in a place for temporary surveillance where power and communications are not readily available can be complicated and expensive....
Testing Bandwidth Vs. Low Light on Jan 16, 2019
Nighttime bandwidth spikes are a major concern in video surveillance. Many calculate bandwidth as a single 24/7 number, but bit rates vary...
Avigilon Favorability Results 2019 on Jan 15, 2019
Since IPVM's 2017 Avigilon favorability results, the company was acquired by Motorola and has shifted from being an aggressive startup to a more...
Gorilla Technology AI Provider, Raises $15 Million, Profiled on Jan 15, 2019
Gorilla Technology is a Taiwanese video analytics manufacturer that recently announced a $15 million investment from SBI Group, saying this...
The False SCMP Story on Hikvision NYC AI on Jan 14, 2019
In the past week, one of Asia's largest publications, the South China Morning Post (SCMP), posted an article about "Chinese [facial recognition]...
WDR Tutorial on Jan 11, 2019
Understanding wide dynamic range (WDR) is critical to capturing high quality images in demanding conditions. However, with no real standards, any...
Pelco Favorability Results 2019 on Jan 11, 2019
Pelco had a significant favorability problem amongst integrators in our previous study (see 2016 Pelco results). Now, in the first edition of our...
CES 2019 Show Report on Jan 10, 2019
IPVM was at CES 2019 for the second year (see our CES 2018 Show Final Report) and is reporting on announcements and interesting news from the show...

Most Recent Industry Reports

The IP Camera Lock-In Trend: Meraki and Verkada on Jan 18, 2019
Open systems and interoperability have not only been big buzzwords over the past decade, but they have also become core features of video...
NYPD Refutes False SCMP Hikvision Story on Jan 18, 2019
The NYPD has refuted the SCMP Hikvision story, the Voice of America has reported. On January 11, 2018, the SCMP alleged that the NYPD was using...
Mobile Surveillance Trailers Guide on Jan 17, 2019
Putting cameras in a place for temporary surveillance where power and communications are not readily available can be complicated and expensive....
Exacq Favorability Results 2019 on Jan 17, 2019
Exacq favorability amongst integrators has declined sharply, in new IPVM statistics, compared to 2017 IPVM statistics for Exacq. Now, over 5 since...
Testing Bandwidth Vs. Low Light on Jan 16, 2019
Nighttime bandwidth spikes are a major concern in video surveillance. Many calculate bandwidth as a single 24/7 number, but bit rates vary...
Access Control Records Maintenance Guide on Jan 16, 2019
Weeding out old entries, turning off unused credentials, and updating who carries which credentials is as important as to maintaining security as...
UK Fines Security Firms For Illegal Direct Marketing on Jan 16, 2019
Two UK security firms have paid over $200,000 in fines for illegally making hundreds of thousands of calls to people registered on a government...
Access Control Cabling Tutorial on Jan 15, 2019
Access Control is only as reliable as its cables. While this aspect lacks the sexiness of other components, it remains a vital part of every...
Avigilon Favorability Results 2019 on Jan 15, 2019
Since IPVM's 2017 Avigilon favorability results, the company was acquired by Motorola and has shifted from being an aggressive startup to a more...
Gorilla Technology AI Provider, Raises $15 Million, Profiled on Jan 15, 2019
Gorilla Technology is a Taiwanese video analytics manufacturer that recently announced a $15 million investment from SBI Group, saying this...

The world's leading video surveillance information source, IPVM provides the best reporting, testing and training for 10,000+ members globally. Dedicated to independent and objective information, we uniquely refuse any and all advertisements, sponsorship and consulting from manufacturers.

About | FAQ | Contact