Masks Cause Major Facial Recognition Problems

By: Rob Kilpatrick, Published 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 systems.

IPVM Image

IPVM tested four facial recognition systems to see how they performed with masks:

The 1-minute video below overviews our findings:

*********** ** ******** ** increase ** *** *** of ******* *****, ***** new **** **** ******* show ***** ***** ******** for ****** *********** *******.

IPVM Image

**** ****** **** ****** recognition ******* ** *** how **** ********* **** masks:

*** *-****** ***** ***** overviews *** ********:

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

Masks *********** ****** *********** ***********

** *** *****, ****** recognition ********** ******* ************ (over ** ****** **********) when ******* ****-***** **** masks, **** ******** *** recognized ** *** ** most *****, *** ** faces **** ******** **** of *** ****.

************'* ********** ********* **** ** ************ recognize ********, *** **** required ********** ** ** decreased **** **** ** medium ** ***. ***** recognition *** ********, ** is ******* **** **** will ****** ** ***** identifications ** ***** *********** with ******** ** ********* of ********.

Further ******** ********

** ** ******** **** these ************* ***** ******* masked **** *********** **** further ******** ** ***** algorithms. *******, *************, ********* will **** **** ***** points ** **** *********** on ** ** **** of *** **** ** covered, ****** *********** *********, even **** ********.

Drastically ******* **********

***** *****, **** ******** to *** **** ******* not ******* * ****, confidence ******* ** **** 60 ****** **** ******* a ******* ****, **** nearly ***% ********** ** just **** **% ** the********* **** **** **** rec ******.

IPVM Image

** ***** ** ********* a **** ** ***** levels, ******* ********** ****** to ** ******* ** ~25%. *******, *********** ******** inconsistent ** **** *******, with **** ******** ********, but *** **********. **** was ********** ****** **** hair ******* *** *******'* forehead, ****** ** **** difficult ** ****** *** face.

IPVM Image

**************** *********, **** *********** reduced ********** *** ******** total ******, **** **** using *** ********** ***** requirements. *** *******, *** face ***** ******* **** only **-** **********, ******* of **-** **** *** same ******* *** *** wearing * ****.

IPVM Image

Avigilon **** *********** **** ******** *** **** ******/*** **********

** *** *****,********'* ****** ************** *** **** *********** rates, *********** *** ******** walking ******* *** ***** when ***** ****** ** low ********** ******** (******** does *** **** ***** confidence ******).

IPVM Image

*******, **** **** ***** recognition ****** ** *** tests ***** * ***** population, ***** ****** *** assume ** ***** ******** similarly ** ****** ************* where *********** *** **** diverse. ** ** ******* that ***** ************ *** missed ************ **** ******** as *** ****** ** subjects ********* ******* **** more ****** **** ** more *********** *** ****** to **** ** ****** to *** ****** ** looking **********.

No ***** ******** ** *****: *******

** *** *****,******* ****** ********************* ****** ***** ** people ******* ******* *** scene **** ******* *****. These ********'* ***** **** not ******** ** *** and *** ********** **** Verkada's **** ****.

IPVM Image

All ******* *** **** **********

*** ******* ****** **** miss ****** ********** - meaning **** *** ****** did *** **** ********* a **** *** *******, which ** ********* ** even *** ** ********** what ******'* **** ** is:

  • ******** ****** ****** ***** walking ******* *** *****.
  • ********* *** *** ****** best *********** **** *** at *** **** ** a ***** ********** ********* of **. *** ******* higher ********* *** *********** misses.
  • ******* ****** ****** ********** and ****** ******** ***** in *** *****.
  • ******* *** *** ********* anyone ** *** ***** wearing * ******* ****.

Improvements ** ****** ********* **** ***** *******

********* * **** **** a **** ** * solvable *******. *** ***** is **** ******* ****** did *** ***** ***** systems ** ********* ***** with *****. **** *** be ****** ** ****** new ******** **** ** faces **** ***** **. As ****, ** ***** expect ****** ******** ******** from **** ******* ** include ****.

Harder ********* ** ********* ***** **** *****

***** ********* *********** ****** go **, *********** *********** faces *********** ******. ***** hide *** ****, ****, mouth, ****, *** ***, essentially ************ ******* ** match ** **** *** forehead.

IPVM Image

****** ** ********* ***** can ***** *** ** be **** *** **** sometimes ** ***** ** correct *** ************* *** accuracy, ********** **** ****** scale, **** ** *** reduced.

*** ****-**** ******** **, structurally, ******** ****** ******* where ******* ***********, ****-********** lighting, *** ****** ********** their ***** ********* *** odds ** ********. *** example,*****'* ******* ********** ** already ********** ** **, ** their ********* ***** ***** demonstrates:

*** **** ** ******** works *** *** **** mistakes *** **** ** not *********. *******, ****** control *** **** ** tuned ** ***** **** matches / ****** ** the ********** ** **** that ** ** ****** to *** ****** ** quickly **** ***** ***** users **** ********.

** ********, ****** *********** in ***** ************, *.*., the ***-***** ***** **** examples *** ***** ** face *** ****** ******** - ********, **** *******, wide *****, ***., **** combined **** **** ***** are ****** ** **** facial *********** ******** *** norm, **** **** *************.

Comments (10)

*** *******, *** **** below ******* **** **** 30-40 **********, ******* ** 60-70 **** *** **** subject *** *** ******* a ****.

* ** *** **** I ***** ********* ***** anymore **** ** ******* a **** :)

****** *** ******* ***** recognize ******** ***** **** masks.

*** *** ********* ****** there *** ** *********** that ****** *********** ***** work **** **** *****?

** *** ********* ****** that ********* *** ******* were ******** **** ** we *** * **** test ** ******.

***, ***** *** ****** on **** **** *** can ***** *** **** think ** **** **** but ** *** **** actual ******* ** **** what *******.

* ******* *** ***** were **** **** *** same **********?

***, **** **** *** done **** ***** ******* (that *** *** **** resolution ********* *** ******** and *******), ****** ** also **** ** *** Avigilon *** *******, *** only ********** ******* ***** and ** **** ***** masks *** *** ******** detected.

*********, ******** **** ***-******* the *******, (*** *******) will ******* *** ****** recognition. **** ** ************ noted ** * ****** Insane ***** ***** ****. (Seriously)

********** *** ****** *********** can **** **** **** masks **, *** **** how **** ** **.

******, #*, ***** **** reports ******* ******** ****** our **** *** *********, which ** ***** ** the **** *****.

*** ******* ** **** accuracy ** ***** ** decline ********** ** **** optimization ** *********, ** these ******* ****.

** **** ***** ************* need ** ***** ** how **** ** *****, at *****, ** *** wild, ***., *** **** that ** *** ************* work.

*********, * ******* ******* called ************ ** **** ************ ********** ** ** 95% ******** **** *** with *****. **** **** claim **** *** ******** everyone ** * ***** of ** ** ** (masked) ****** “****** * second”. ****** **********.

Login to read this IPVM report.

Related Reports

Clinton Public View Monitor (PVM) Mask Detection Tested on Jul 09, 2020
Face mask detection, or more specifically not wearing one, is expanding...
Faked Coronavirus Fever Detection, Athena Used Hikvision; Responds - Selling NDAA Compliant Cameras, Pledging 50% Of Profits to Victims on Mar 24, 2020
US company, Athena Security, faked its coronavirus fever detection marketing,...
Facial Recognition: Weak Sales, Anti Regulation, No Favorite, Says Security Integrators on Jul 07, 2020
While facial recognition has gained greater prominence, a new IPVM study of...
Integrated IR Camera Shootout 2020 - Avigilon, Axis, Bosch, Dahua, Hanwha, Hikvision, Panasonic, Uniview, Vivotek on Jan 30, 2020
The best and worst cameras tested in this IPVM shootout showed major...
Facial Recognition 101 on Mar 18, 2020
Facial recognition interest, use and fear is increasing. This guide aims to...
Avigilon Face Mask Detection Tested on Jun 24, 2020
Face mask detection or, more specifically not wearing a face mask, is an...
Fever Camera Sales From Integrators Surveyed on Jun 01, 2020
Fever cameras are the hottest trend in video surveillance currently but how...
Anixter Runs Fake Coronavirus Marketing Using Shutterstock Watermarked Images on Jul 24, 2020
Coronavirus faked marketing is regrettably commonplace right now but Anixter...
Smartphone Thermal Camera Body Temperature Measurement Tested (FLIR / Seek) on Apr 10, 2020
Thermal cameras have exploded in popularity due to coronavirus, but can...
FLIR Screen-EST Screening Software Tested on Jun 30, 2020
In our FLIR A Series Test, the cameras' biggest drawback was their lack of...
Dahua, Hikvision, ZKTeco Face Mask Detection Shootout on Jun 19, 2020
Temperature tablets with face mask detection are one of the hottest trends in...
Injes Tiny Temperature Terminal Tested on Jul 17, 2020
While temperature terminals have trended bigger, the Injes DFace801 is...
Coronavirus Hits Manufacturers, Standing Now, Worse To Come on Apr 06, 2020
Coronavirus is hitting security manufacturers, though overall modestly for...
The Problem With Fever Detecting Thermal Sunglasses on Apr 15, 2020
While the media has promoted using thermal sunglasses to detect fevers, this...
TVT / InVid Facial Recognition Tested on Mar 25, 2020
Facial recognition is frequently sold for thousands of dollars per channel...

Recent Reports

Genetec CEO Declares "We Don't Negotiate Payment With Patent Trolls" on Aug 11, 2020
Are patent trolls like terrorists? Genetec's CEO is coming out strongly...
Hanwha AI Analytics Camera Tested on Aug 11, 2020
Hanwha has released their Wisenet P AI camera, adding person and vehicle...
Alabama Schools Million Dollar Hikvision Fever Camera Deal on Aug 11, 2020
The Baldwin County, Alabama public schools purchased a $1 million, 144-camera...
Dahua Taunts Australian Government, Continues To Sell Illegal Fever Cameras on Aug 10, 2020
Dahua is effectively taunting the Australian government by continuing to sell...
HID Releases VertX Replacement Aero on Aug 10, 2020
HID is replacing two established and broadly supported types of access...
NDAA Compliant Video Surveillance Whitelist on Aug 10, 2020
This report aggregates video surveillance products that manufacturers have...
Telpo China Temperature Tablets Tested on Aug 10, 2020
The provider for overseas companies ranging from Canon Singapore to US'...
Dangerous Hikvision Fever Camera Showcased by Chilean City on Aug 07, 2020
Deploying a fever camera outdoors, in the rain, with no black body, is...
"Grand Slam" For Pelco's PE Firm, A Risk For Motorola on Aug 07, 2020
The word "Pelco" and "grand slam" have not been said together for many years....
FLIR Stock Falls, Admits 'Decelerating' Demand For Temperature Screening on Aug 07, 2020
Is the boom going to bust for temperature screening? FLIR disappointed...
VSaaS Will Hurt Integrators on Aug 06, 2020
VSaaS will hurt integrators, there is no question about that. How much...
Dogs For Coronavirus Screening Examined on Aug 06, 2020
While thermal temperature screening is the surveillance industry's most...
ADT Slides Back, Disappointing Results, Poor Commercial Performance on Aug 06, 2020
While ADT had an incredible start to the week, driven by the Google...
AHJ / Authority Having Jurisdiction Tutorial on Aug 06, 2020
One of the most powerful yet often underappreciated characters in all...
SIA Coaches Sellers on NDAA 889B Blacklist Workarounds on Aug 05, 2020
Last month SIA demanded that NDAA 899B "must be delayed". Now that they have...