BriefCam Video Analytics Tested

By: Ethan Ace and Rob Kilpatrick, Published 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?

IPVM Image

We tested BriefCam for over the past month to see how it performed, examining:

  • How does BriefCam search compare to typical VMS search?
  • How well does clothing style and color analysis work?
  • Does it accurately classify age and gender?
  • How well does facial recognition work?
  • Are vehicles accurately classified and searched?
  • How well does mobile app search work?
  • How does it compare to competitors such as Avigilon and Anyvision?

********,******** ** ***** ** 2018, ** *** ** the **** ******** **** video ********* ********* ** the ****. *** *** well **** ** ****?

IPVM Image

** ****** ******** *** over *** **** ***** to *** *** ** performed, *********:

  • *** **** ******** ****** compare ** ******* *** search?
  • *** **** **** ******** style *** ***** ******** work?
  • **** ** ********** ******** age *** ******?
  • *** **** **** ****** recognition ****?
  • *** ******** ********** ********** and ********?
  • *** **** **** ****** app ****** ****?
  • *** **** ** ******* to *********** **** ** Avigilon *** *********?

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

*******

******** ****** ******** ***** in *** ***** ******** platforms, **** ***** ***** synopsis *** ******** ********* speeding ****** ************* ******** to ******* ***, **** features **** ** *****, clothing *****, ******* ****, direction, ***., ******* ************ well ** ******* *** moderate *** ***** ******, useful *** **** ********* and **** ******. ************, its ****** *********** ********* were *********** **** ****** we **** ******, **** as ********* *** ********, with ** ***** ************ in * ***** ** testing.

*******, ******** **** ** age *** ****** ******** were **********, **** *** detected ** ***** *** adults ******** ** ********, and **** *****. ************, frequent ****************** ** ********* objects (****** ***** *** large *******) ** ****** will ****** ****** **** analytic *********** ** ******* scenes.

*******, *** *****, **********, and ****-*** ******** ************ (at ***** *** **** required) *** ** *** of *** ***** ** many ***********, ****** ** most ****** * *** in ****-*** **********, *********, or ********** ********, ***** is ***** *** ******* has ************ *******.

Vs. ******** ********** ******

******** *** ***** *** advantages ******** ** ********:

  • ***** ********:********'* ***** ******** ** its **** ********* **** Avigilon's ********** ******, ** it ****** **** ****** review ** ******** ***** compared ** ****** ******** or ********* ******** ******** in ******** ******* ******.
  • ***** **** ** ******:************, ******** ****** ******** faces *** ********** ** be **** ** * search **********, ***** ******** only **** ****** *********** in **** ******.
  • **** ********:******** ********** ****+ ******* ******** *** ******* video *** ******* ****** (including ******* *** *********) for **** ******. ********'* Appearance ****** *** ****** recognition *** ****** **** with ***** *** *** VMS.

*******, ******** *** **** advantages **** ********, ** well:

  • **** ******** ****** *********:***** ******** ******** **** false *************** ** ********* objects ** ******,********'* ****-******** ************* **** **** ******** in *** *****, **** few ***** ****** ** animals, *******, ** ****, making **** **** ********* useful *** **** ******.
  • ***** ********** ******:************, ********'* ********** ****** is ***** ********** ** its ***, ** ***** may ****** ******** ******** directly **** **** ** recorded *****. ** ********'* integrated *****, ******** *** performed ******* ******* *** embedded ******** ******.

Vs. *********

** ***** ** ****** recognition ********, ******** *** Anyvision ********* ********* ** our *****, **** **** few ** ** ***** recognitions (*** ******* ******** for *******). ********* *** moderately **** ******** **** subjects **** *******, *****, or ***** ***********, ****** BriefCam ********* ****** ** WDR *** *** ***** scenes.

*******, **** **** ********* does *** ******* *** video ******** ** ******* object ******* ** ********, such ** ********, *****, direction, ***, ***** **** BriefCam **** ********* ****** for ******** ******.

Key *********

******** *** ******* ********* in *** *******:

  • ******* ****** ****:********'* ***** ******** *** forensic ****** ******* *** time ****** ** **** specific ******* ** ******** in *** ****, ** the *********** ** ****** video ****** *** ****** filters ****** ******** ******* from ******.
  • ******** ****** ********* ****:****** **** **** ** color, ******** *****, ******** detection, ********* ** ******, object ****, *** ****** accurate ** *** *****, features *** ***** ** most, ** ***, *********** analytic *********.
  • ***** **** ***:********'* ****** *********** *********** was *********** **** ******, as ****, **** **********************'* ********** ******, **** ** ***** recognitions ** ***** ***** of *******, ****** **** moderate ***********, ******** *****.
  • ******** ******* **************:******* ****** **** *** very ******** ** ****, identifying ***** *** **** (car, *****, ***, **********, etc.) **** **** *** misclassifications.

*******, ******** ******** **** multiple *********, ** ****:

  • ********* ******* ********** ** people:** *** *****, ***** animals, **** ********, *** blowing ***** **** *** regularly ********** ** ******, regardless ** ******** ** optimize ********.
  • ********** ***/****** ********:****** **** ********** ** children *** ***** ** men (*** **** *****) in *** *****.
  • ******** **** ********: ******** ******** * minimum ** ** ***** two **** ********** ****, as **** ********* *** synopsis/forensics ****** ** *** simultaneously ** * ****** GPU.

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

***** ********'* ***** ******** significantly ******* *** ****** of **** ******** ** review ***** ********** ** video *** ******* ******* the **** ******** ** find ******* ** ******** across *** ** **** cameras.

** **** ** *** synopsis ** *******, ***** with ****** ******* ***** may ** *******, ** this *****:

Most ******** ******** ********

** *** *****, ***** analysis *** ********* ********. Searching *** ******** ******* red **** ******* ** accurate *******, ****** **** trial *** ***** ** selecting ***** *** ** required.

IPVM Image

************, ***** *** **** sleeve ********* ****** ******** to ****** **** ******* further ** ******** *****, e.g., *** "***** ***** sleeves" ****** ***** ****:

IPVM Image

************, ********* **** ********* accurately ********, ***** ****:

IPVM Image

Age *** ****** **********

****** *** *** ************** proved ** ** *** most ********** ************** ** our *******. *** **** long **** **** ********* classified ** *****, ***** below.

IPVM Image

************, ****** **** ******, many ****** **** ********** as ******** ** *** tests.

IPVM Image

***** *************** **** ********* inaccurate ** ********'* ********** Search, ** ****.

Accurate ******* ********

******** ************** ******* ******* types, *********: ****, *******, trucks, ****, *****, ******, boats, ********, *** ***********.

IPVM Image

******* ******** *** ********, as ****, **** ** the ***** **** ***** below.

IPVM Image

*******, ******* ***** *** inaccurate ** *** *****, with ********** *********** ******* many **** ** ****** appear ** "***."

IPVM Image

Direction / **** / **** ********

******** **** ****** ***** to ****** **** ********/****** by ****** *********, * drawn ******** ****, ** area *******, *** ** which **** ******** ** our *****. *** *******, the **** ***** ***** pickups ********* ** *** right ** *** ***.

** ******** ** ****** direction, ***** *** **** draw * ******** **** on *** ******'* ***** of ****, ********* ****** to **** ***** ****** along **** ****. ** our *****, **** ****** accurately, *** ***** ** useful ** ****** ***** only ** ***** ******* which ******** **** ***** path.

** *********** **** ******* in **** *****:

Natural ******** ****** *** ******

** ******** ** *** main *** ****** **, BriefCam **** ****** * mobile *** ***** ****** users ** ****** *** objects ***** ******* ********. For *******, * **** may ****** **** "****** in *** ** **** hour" ** ***** * search. ******* *** ******** in *** *** *** also ****** ** * case ** *** ****** for ***** *********.

** *********** **** *** in **** *****:

**** ******* *** ************* is ********, ******** * simple *** ** ******** searches **** * ****** device ******* ******** ** app's **.

*******, **** **** ***** searches *** *** *******, but *** ******* ****** for **********. ********* *** days ** ***** **** ~10 ******* ** *** testing, **** *** *** notifying ** **** ** was ********. ******** **** appear ** ********'* **** UI ** **** *** be ********* *** ********* from * ********* ******.

No ********** ***** **********

***** * ******** ** over *** ****** ** over ***** ** *******, no ******* *** ******** for ******* ****** ** the ***** ****.

** * **** ***** setting, ***** *** ********** detected ** ~**-** *** or ~**-** ****** *** face, **** ** ***** than ********* ***** ******** ~85-90 *** *** ******** Facial *********** ***** ******** ~90 ***, **** ******* lower **** **** ***** platforms ********* ********* ~*** PPF *** ******** ***********.

IPVM Image

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

** * **** ***** setting, **** *********** ***** not ***** * **** to *** ***** **** at * *' ****** and ~**° ******** **** the ******* ********** ** an ***** (~**° ***** of *********).

IPVM Image

** ******* ******, *********** was *** ********, **** as *** **** **° angle ***** ****. **** that **** ****** *********** platforms ** *** ****** at ***** ******, **************'* ********** ********* **** **.

IPVM Image

Performance ********* ** *** ***** *** ****/**********

****** **** ***** **** to ** ********** ** low ***** ** ****** camera ******* ** ~**' and ********* ** ~**° but **** **** *** lowest ********** *******. *********** was **** ******** **** low ********** *********** ** a **° ***** ** a *** ****** ****** of *' *** **° downtilt.

*********** **** *** *** similar ** * ****/********** setting ** ~*.****, ****** camera ******* ** ~**' with ~**° ** ******** resulted ** *** ********** detections *** **** ***** camera ******* ** ~*' with **° ******** **** recognized **** *********** ** a **° *****.

IPVM Image

Low ********** *********** **** ***/**** *** **********

****** *******, **** ** a *** ****** ****** of ~*' **** **° of ******** ****** ******* wearing *********** **** ********** parts ** *** **** (e.g. ********** *** * hoodie, ***** *****) **** only ********** **** *** lowest ********** *******.

IPVM Image

People ******** ******** ********, *******, *** ********

****** *******, ****** **** properly ******** ** ****** walking ****** *** *** night, ** **** ** when *******, ********, ** partially ********, ***** ****:

IPVM Image

People ******** ** ~* ***

** *** *****, ****** were ********** ******* ** at ~* ***, ***** a ***** ****** **** a ***° ***** ** view **** ******** ** a ~***' ********* ********. This ** ******* ***** than ******* *** **** performing ********* **** ** Avigilon's ** (~** ***) or ***** *** (~* PPF).

IPVM Image

False ****** ****** ** *********, *******, *** *******

********* ******** **** *** camera's **** **** ********** classified ** ******, **** when ***** ****** ********** requirements.

IPVM Image

******** ** **** ******** in *** ******** **** stamping **** ***** ** the **** *** *********** them ** ******, ***** visually ********* *** ******** during **** ******, ****** it **** ********* ** review.

***** ******* ******* ******* the ***** **** ** deer ** ***** **** were ********* **** ********** as ******, ***** ********** of *****/******/****** ********** ********.

IPVM Image

*******, ******* *******/***** *** classified ** ******, ** well, *** ***** ***** not ** ******* ** using ******** ********.

IPVM Image

BriefCam ******** ** ******************

**** ***** ***** ***** false ***************, ******** ********* that ***** ******** ** not ******* ** ****** them. *******, ***** ******** is ** ******* ***** objects *** "****** *********" to ********* ** **** may ********* ** ******. Additionally, **** ********* **** events *** ****** ** relevance ** ******* *** these *************** ********* *** not ***** ** **** confidence, ** ***** ** the ******* ******* *****.

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

******** ** ***** **** to ******* ************** **** large ********** ** ***** ** ** ******* months **** ******** *******. As * ******, ***** conduct *** ************* ** reviewing *** ******** ******* via ***** ********® ** thumbnail **** ****** **** viewing *** ******** *****. Therefore, ** ** ******** that ** ** *** miss *** ********* ** any ******* ** *** scene. ******* **** ** are ****** ** ******** correctly (******* ******* ** have *** ******* *** DNNs **** ******* ****) are *** ********* ** ensure ** ** *** miss ********* ******. *********, an ************* ************ ** obtained ** *********** * subtle ******** **** ** hardly ************ *** ************ by *** ***** ***; raindrops **** **** **** category. ******** ************ ******* many ***** ** ****** objects **** ** ****** leaves, ***** *********, ***** graininess *** ******.

*******, **** ** ******* to*********'* ******** ** ******* inanimate ******* ***** **** detected ** ***** ** operators, ******* ** ******** every *** ********** *********, as ***** *** ** removed ***** ***** ********* be ****** ** *************.

VMS ****** *********** *******

********'* *********** ** ***** is ****** ******* ** reproducing *** ******** ** in *** *** *** or ******** * ******** to ** ******** **** VMS *******, ******* ** integrating ****** ******** *** search ********* **** *** VMS ****** *******.

******** ***** **** **** to **** ***** (*******, Indigovision, *** *********), ** any **** **** ** smart ***** *** ** used ** ******* ** alarm, ***** **** ** object ********, **** ** a **** ** *******. However, ******** ****** ******** such ** *****, *********, speed, ***., ** *** sent ** ****** ** the ***, *** ******* available ** *** ******** client.

** ****** ********* ******** Smart ****** *********** ** this *****:

** ******* ** * Milestone *****:

IPVM Image

***** ****** *********** ** limited, *********** ***************** ** ******** ****/******** video *** ******* **** other ********* ** **** tested, **** * ****** installed ** *** ******** server, *** *** ****** credentials *******, *** ****** other *****.

IPVM Image

High ******** ************

****** **** ******** ********* we **** ******, ******** requires ** ***** *** GPUs ** *** *** full ***** ** ********, as **** **** ** dedicated ** ****** ****-****/**** alerts ** **-******/***** ******** processing. * ****** *** cannot ** **** *** both.

**** **** ********** ******** to ****** ***** ******* processing ** * ****** GPU. ******** ******** ***** cost **** **** ******, such ** *** ****** GTX ****, ***** ***** for ~$*** *** ******, which *** ****** ******** cost *** **** **** for **-****** **********. *******, for ***** *******, **** expensive ****** *** ***** GPUs **** ** ****, similar ** ******** *** Anyvision ************.

******** **** *** ******* exactly *** **** ******* a ***** *** *** handle, ** ** ******* on **********, *********, *** bitrate. *** *** **** in *** **** *******, the *** *** ** BriefCam's ************, *** **** to ******* **** **** alerts *** **** ~*-* 1080p *******.

*******

******** ***** ** ******** only *** ~$***-*** *** per *******, **** * 100 ******* ******* ***** (~$25,000 **** ****). ************, at ***** *** **** are ******** (*** *** real **** ******, ******* for ******** **********), ****** ~$2,000+ ** ********.

*** ******* *** (~$**,*** USD **** ****) ** available *** ******* ******* counts, ***** ******** ****** hardware *** *********.

Comments (17)

*** ****** **** **** gender ********* ******** ********? I **** ****** ** inside ** ******* ********* products *** **** ***** seen ** ******* *** where **** ******. ** seems **** ***** **** either ****** ******** ***** or ******** ******* **** and ******* ***** ** no *** *** ****** recognition ** ** ******* than **% ********. * honestly ***'* ********** *** gender *********** ** ******** given **** ** ***** seems ** ** **** moderately ********... **** **** feature ***** ** ********* the ***** ******* *** this ** *** ****** to ********.

**** **** ** **** that *** *******'* ****** gender. :-)

** ***** **** ** a *** **** *** gender ********* ** *** security *****? ****** *********, in ** *******, *** going ** ** **** for ********* *** *********** purposes. *** ** **** security.

**’* ***** ** ** interesting ** *** *** these ***** ***** ******** plugin ********* *** ***** to ****** **** *** next * *****. ********* like ********, ***** *** Verkada *** ********* * lot ** *** **** technology, ** * ******** offering, ** ***** ***’* for **** ****, **** hardware, **** *****...***.

** ***** **** ** a *** **** *** gender ********* ** *** security *****?

****** ****: "** ** the ******* *** * white ****, *** *-*, wearing * ***, ***** jacket, ***." **** **** raises *** **** ** using ****** ** ****** analytics, ****** **** *** AI ********* ***** ********* the **** ********* **** - ********, *********, ********, offer. *******:********* ******* ****** ********* Analytics, *** ****** **

**** *****......***, *** ******* in **** ******** ***** the ********* * ****** chance ** ******* *** perp **** ** *** estimation **** ****. ** think *** **** *** marketing *** ********** ** where *** ***** ** at?

** ***** *** **** for ********* *** ********** is ***** *** ***** is **?

* ***** ***** ** 'money' ** **** *****. Generally, ************ ******* ******* have *** ********** ******** and ******* ********* *** marketing *** **********, ** they **** ** **** different ******** *** *** through ********* ******/********.

@**** **** *** **** test **** ******* ***********? It ***** **** **** tests ** **** *** done **** ********* *****. I'm ******* ** ******* for ***, ****** *** or ***/***** ********* **** vary ********* ** *** persons *********.

******, ** **** * general ********* ** *** to **** **** *********** numbers ** ****** **** privacy ********, ** ** we ***'* **** **** variety ** ********* *** more ******* ******* ** all. **'* ********* ** are ****** ** ****** out *** ** ** without ********* *******, ***.

************, ****** **** ******, many ****** **** ********** as ******** ** *** tests.

*** **** **** *** the *********** ********* ****** installed ;-)

** ***** ** *********** to ********** *** **** streams ****** *** *** handle ************** *** **** type ** **** *** supported ** **** *******.

******** ******** **** ****** to ** ***** *********** where *** *** ***** project **** (****** ** cameras, **********, ******* ****, etc.) *** *** ******** suggestions. *** ****** ** required **** ** **** of *** ********** **** is *********.

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

"******** ******** ** ***** two **** ** *** the **** ***** ** features, ** **** **** be ********* ** ****** real-time/live ****** ** **-******/***** synopsis **********. * ****** GPU ****** ** **** for ****."

"******** **** *** ******* exactly *** **** ******* a ***** *** *** handle, ** ** ******* on **********, *********, *** bitrate."

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

**** *** *** ********* processing **** (***) ************ for *** ******** ********? | ********

**** *** *** ********* processing **** (***) ************ for *** ******** ********?

*** ******** ********** ****** requires ** ****** *** to ******* *****. *** exact ***** *** ****** of *** ***** ******** will ****** ** *** requirements ** *** ********** of ***** ** ** processed ***** **** *** number ** ***** ** video ** ******* *** day.

*** * ****** **** a ***** ***** ********** requirement, ******** **** *** be ********* ** **** processing ******, ** **** as *** ******* ** utilize ******** ********** ******* simultaneously.

**** *** ********* ** the ******** ****** **** be ******** ** ******* video ** * ******** processing ****, ****** *** on-demand ********** (****** *** RESEARCH *******) ** *** real-time ********** (******* ******), a ****** *** ****** process ***** **** ****-**** and **-******.

*** ******* ****** *** can ***** *** ** cameras (*** **** *** storage **** ** ********)?

*'* ****** ** **** reached *** ** ******** to *** ** ******* that *********** ** * don't **** ** *********** from *** ******* **** is ** ** ********** confidential/"eyes ****" ** ***.

******* **** ** ******** in ******* **** *********** hardware ************ *** ******** as ** ******** ** how ******** ******* *** stores ***** *** ******** when ********** ** *************/********, but ********* *** ******** for **** *********** *** needs.

* *** ****** ** benefits ****. **** ***** like * **** ********* product.

**** **** ** * not ****** ** ******* here? **** *** *** reasons ** **** ** another *******? (******** ** out ** *** ***** range *** ** ***** client.)

**** ** ******** ** the ******* *** *** $20k? *** **** ********? Can ** ** ******** upon, *** **** ********, more ****, **** *******?

***, * ********* ******** about ****** ******* *** they ******** ** **** the ******* *** (***** are ******** ** *********/*******) includes * ********** ***** and ** ****** ******** which *** ** ******** to *** ********. ** far ** ****** **** GPUs ***** ******** ** below.

*** **** ****** *********** be **** ** *** an ********** ****** *** 4000 *** **** **** server ** ********** *** might **** **** ** increase *** *** ** 128GB ********* ** *** expected ********** **** ****.

** ***** * ******* that ****** ****** ******** that *** ** ******* or ************* (***** *****). Ie. *** ***, *** clothes *** **** ****** as **** ********* ** to ** ** ***** in **** ** *** subjects ** **** ****.

* ***** ***** ****** for *** ** ********, and **** *** ******* where ** *** ******** says "*******."

Login to read this IPVM report.

Related Reports

New Axis M30 Cameras Tested on Mar 26, 2020
Axis has released a new generation of, for them, relatively low cost M30 series cameras, claiming to deliver "sharp video quality even in poor...
FLIR Screen-EST Screening Software Tested on Jun 30, 2020
In our FLIR A Series Test, the cameras' biggest drawback was their lack of face detection, requiring manual adjustment when screening each...
Avigilon ACC Cloud Tested on Jul 08, 2020
Avigilon merged Blue and ACC, adding VSaaS features to its on-premise VMS, offering remote video and health monitoring that was previously limited...
Uniview Deep Learning Camera Tested on Jul 14, 2020
Uniview's intrusion analytics have performed poorly in our shootouts. Now, they have released their Prime Series Deep Learning cameras, with an...
Camera Analytics Shootout 2020 - Avigilon, Axis, Bosch, Dahua, Hanwha, Hikvision, Uniview, Vivotek on Jan 22, 2020
Analytics are hot again, thanks to a slew of AI-powered cameras, but whose analytics really work? And how do these new smart cameras compare to top...
Video Surveillance Cameras 101 on Feb 25, 2020
Cameras come in many shapes, sizes and specifications. This 101 examines the basics of cameras and features used in 2020. In this report, we...
Video Surveillance Business 101 on Mar 30, 2020
This report explains the fundamental elements of the video surveillance business for those new to the industry. This is part of our Video...
VSaaS 101 on Mar 25, 2020
Video Surveillance as a Service (VSaaS) is the common industry term for cloud video. But what does it mean? How does it all work? Inside this...
YCombinator AI Startup Visual One Tested on Apr 02, 2020
Startup Visual One, backed by Silicon Valley's powerful Y Combinator, aims to be "Your 24/7 Watchman" with advanced analytics and object...
FLIR A Series Temperature Screening Cameras Tested on Jun 04, 2020
FLIR is one of the biggest names in thermal and one of the most conservative. While rivals have marketed fever detection, FLIR has stuck to EST...

Most Recent Industry Reports

Camera Course Summer 2020 - Last Chance on Jul 15, 2020
Thursday, July 16th is your last chance to register for the Summer 2020 Camera Course. This is the only independent surveillance camera course,...
Bias In Facial Recognition Varies By Country, NIST Report Shows on Jul 15, 2020
While many argue that face recognition is inherently racist, results from one of the most extensive studies done on demographic bias in AI, the...
Video Surveillance 101 July Course - Last Chance on Jul 15, 2020
Friday, July 17th is your last chance to sign up for the July 2020 Video Surveillance 101 Course. This 2-day course is designed to help those new...
Brazil Assembly Powers Hikvision Local Expansion on Jul 15, 2020
Hikvision has grown considerably in Brazil ever since the 2017 opening of an assembly operation in a free-trade zone in the middle of the Brazilian...
50+ Security Industry Companies Take $40 Million PPP Funding on Jul 15, 2020
50+ security industry companies have taken more than $40 million in PPP funds, IPVM has confirmed from US government records. Inside this...
Uniview Deep Learning Camera Tested on Jul 14, 2020
Uniview's intrusion analytics have performed poorly in our shootouts. Now, they have released their Prime Series Deep Learning cameras, with an...
Drako's Companies (Brivo, Eagle Eye) Take $4+ Million in PPP Funds on Jul 14, 2020
While centimillionaire Dean Drako is the owner of two of the largest SaaS businesses in the security industry (Brivo and Eagle Eye), Drako's...
Defendry Presents AI Active Shooter Security System on Jul 14, 2020
Defendry presented its Active Shooter security system at the May 2020 IPVM Startups show. A 30-minute video from Defendry including IPVM...
Ban Rules Released: Use Dahua or Hikvision, No US Government Contracts on Jul 13, 2020
The US government has released the rules implementing the "Prohibition on Contracting with Entities Using" Dahua, Hikvision, and Huawei based...
JCI Sues Genetec For Patent Infringement on Jul 13, 2020
Surprisingly, security giant JCI has sued their partner, security software developer Genetec, for patent infringement. Inside this note,...