Avigilon UMD / UAD Tested

By Rob Kilpatrick, Published Oct 14, 2020, 10:05am EDT (Research)

********'******** ******** ********* *** ******* ****** Detection***** ** "******* ****** **** *** go *********", *** *** **** **** claim ***** ** ** **** ***** scenes?

IPVM Image

** ****** ******* ******** ********* *** Unusual ****** ********* *** **** * month, *********:

  • ** **** **** ***** ** ****** unusual ******** ** ******?
  • ** **** **** ******* ******** ** the ****?
  • ** ****** *** ** *** ***** alert?
  • *** **** *** *** *** *** to *****?
  • *** **** ** ******* ** *** identify ***** ******** ** ** ********* the ******** *******?

*******

***** ** *** *******, ******* ******** Detection (***** ************'* ** *********) ********** ******* ** ******** ******** in *** ****** ******, ******* *** out, **** **** ********** ***** ****** due ** ***** ** ****** ***** detected ** **** ** *** ** detection ** ****** ******** ** *** tests.

*******, ******* ****** ********* (***** **** pixel *** **** *** ** *** most ******* ******** ******* ******* ****) suffered **** *********** ***** ****** ****** by *******, **********, *** ****** ***** turned ** *** *** ** *** testing, ****** ****** ** *********** ************ in ***** ****** *********, ***** ** is ***** **.

*******, ******* *** *** *** ****** users ** ****** **** ******** ****** as *** *******, ** ********** **** are ******* *** ********** (**** ** emptying ***** **** * ******** ******) will ******* ***/*** ******.

Avigilon ********

******** ******* **** ******** ** ***/*** performance:

***** *** *********, *** *** ***, are ******** ** ******* **** ******** events. ***** *** *** ****** **** an ******** ****** ****** ** ***** for *** ***** *** ***** ** “soft ******” ** ************* *** **** are ********* ** *** ***** ** Attention ********* (***). ** *** **** there ****** ***** ** * ****** in * ****** **** **** *** up ** ‘****** ** ****’ **** ** ** ********** *** ******** and ***** **** ** **** *** false *********.

******** ******** *** ***** ** ********* interface ** *** ******* ***** ** deliver ******* ******. ***** ****** *** unanticipated *** ********* * ****/***** ** not ******* ** *******. ** ***** above ***** ****** **** ** ** more ******** *** *** ***** ** be ******** ******* ** ** ******** in *** *** *********. ***** *** the ***** ***** – *** ** providing **** **** ********* (*** *** ability *** **** ** ******* **** cameras).

Unusual ******** ********* **. ******* ****** *********

*** *** *** **** ******* ****, but *** **** *** **** ******** analytic **** ***** **********'* ** *********(*.*., ***, *****, ****, ***************) *** determining **** ** *******, ***** *** uses ***** ****** ********* ******, ***** are ** ****** **** ********. ** our *******, ***** *****, *** *** significantly **** ******** **** ***.

******* ******** ********* ** ********* **** on*** ******** *******. *** ** ********* ******,***,***,****, ***** **** ****.

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

******* ******** ********* ********** ********* ** our *****. ** *** ******* *****, it ********** *** ***** ** ******* (with * ****** ******** ***) ** they *** ****** ****** ** **** area.

IPVM Image

*** **** ********* ** ******* *****. In ** **** ***** ***** ******** normally ****, ** ***** *** ******* subject ***** ** *******.

IPVM Image

*******, ****** ******** ** ******* ***** of *** ***** **** ******* **. In *** ***** ***** *****, ******** frequently **** **** *** ****, ******, to *** ******* *** ** *** right, *** ****** ***** *** **** near *** ****** ****.

IPVM Image

People ******** *** ********** ******* ********

***** *******/**** ***** *** ******** ********, people ******** ** **** ***** ******* the ***** **** *** ********** ** unusual.

IPVM Image

Vehicle ********* ******** ***** ************** ***** ***** *** ******

** **** *********, ***** ** * truck **** ********** * *** ** an *** ***** *** **** ******* on ** ******* ********.

IPVM Image

************, **** ****** *** ********** * person *** ******* ** ******* ******** event. **** ******** **** **** ****** several ***** ** *******.

IPVM Image

Normal ******** ********* ********** *******

****** ******** **** ** * ***** driving ******* *** *** ****** *** day *** ******* **** ********* ********** unusual ****** *** ********** ******* *** that **** ** *** *** ** area ** *** *****.

Unusual ****** ***** ***********

******** ****** ** ** **** *** correctly ******* ** **** ** * person ****** ** * *** ******* area ** *** *****.

False *** ****** ** ******* *** *** **********

*******, ***** *** ****** ** **** accurate ***** ****** *********, ** ******** from ******** ***** ****** ****** ** everyday *********** *** **** ** ***.

*** *******, **** ********** ************ ***** of *** ***** ***** ** *** typically *** ****** ********* *** ******.

IPVM Image

** *** ******* ***** **** ** these *****:

IPVM Image

*******, ***** ***** ******* (*.*., ******* lights ** *** ***) ****** *** events.

Significant **** ****** ** ***** *****

**** ******* ******** *** ******* ****** need *********** ******* ** **** ** learn *** ***** *** ***** ****** accurate ******. *** ***** ***** * week, ***** *** *** ******* ** to *** *****. ****** *** ***** is ******** *******, ****** *********** **** as ******** ******* ******* *** ***** below *** ******* ***** ******:

******* ** ********'* ***** ****-******** *********, UAD/UMD **** * ******** *** ***** shows ******* ******** ********:

IPVM Image

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

**** ******* ****** *** ******* ******** learn ******** ** *** ***** **** a *** ****** ** ****. ***** the ***** ** ******* *** **** will **** *** ****** ** **** both ********* **** *******, ***** ** no *** ** ***** *** ********* to *** ******* ****** ** **** it ********* ******* *** *** **** does ***, **** ** ****** ******** trash **** *** ******** ***** *****.

IPVM Image

Focus ** ********* *********

***/*** ****** *** ********** ** ********'* Focus ** ********* *********, **** ****** highlighting *** ********** ******'* *** ***** blue. ********* *** ****** ***** ** view *** **** ***** ** ****** mouse **** ** **** **** *****.

** ***** **** *********** ******* ****** and ******** ********* ****** ** **** visibly ********* **** ******* * ********* event **** **** * *****-****** ****** because, ** ******* ** ******** **** *******, *** ****** *****/******** *** *** organized ** * *** **** ******* the ******* ********* ** * ******** location. ********, ***** ******** ****** **** facial *********** *** *****-***** *** **** as ***/***, ***** ****** ** *******.

Simple ***** / ***

**** ****** ***** ** ******** *** UAD/UMD. ** *** *******, *** ** enabled *************. ** ******* ********** ***, users ****** ****** ** ** **** a ********.

****** *** ***** ** ********'* ***** Of ********* *********, *********** ** * lighter **** **** ******* ****** *** detected. ****** ** ******** ***** *** be ***** *** ***** ******, ******* to ***** ******** ******, ** *** be ********* ** *** ******** *** faster ******.

*******

******** *** *** *** *** ******** with ********* ** ******** ******* ******, no ********** **** *** ********.

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

*** ********* ******** **** **** ****** testing.

  • ******** ******* ******: *.**.*.**

Comments (4)

Additionally, this spider was considered a person and created an Unusual Activity event.

Happy to see our Spider Man detection is working!

In all seriousness, I am curious about the truck scene highlighted. Do you typically see trucks in your parking lot get that close to parking spots/drive across the end of the dock ramp in that manner?

That's my initial guess for what may have triggered it, especially as the alert stopped a couple seconds after that. Reminds me of our initial release that we showed at ISC West a couple years ago with UMD: triggered on a person (again, UMD, so this wasn't classified) walking in a park and dropped in and out, seemingly at random.

Turns out, the reason it triggered was snow covered the path and the camera remembered the areas of the path. So it would alert and drop accordingly as the person walked through the snow.

Great test. Enjoyed reading through it.

Agree: 1
Disagree
Informative: 1
Unhelpful
Funny

We actually have trucks drive pretty much that exact route every day. The company next door does distribution so there is truck traffic frequently from late afternoon to the middle of the night. It's possible it could be closer to the dock or building, though, but it appears pretty typical. I'm also wondering if it would become more "normal" given more time, if the camera learned more routes, etc.

Agree
Disagree
Informative: 1
Unhelpful
Funny

I would imagine the answer would eventually be yes, but what I'm unsure of is how many instances it would take for the event to register as normal.

User feedback/training on UMD/UAD is high on my wishlist for these features. I think that's really what will allow it to be fine-tuned.

Thanks for the feedback!

Agree: 3
Disagree
Informative: 2
Unhelpful
Funny

Ryan, Send me a message I would be happy to chat about UMD and UAD.

Agree
Disagree
Informative: 1
Unhelpful
Funny
Subscribe to IPVM Research to read the full report.
Why do I need to subscribe?
The IPVM Research Service includes products tests and shootouts plus competitive and financial analysis, helping decision-makers better evaluate purchasing, partnering, developing, and/or competing against companies in physical security.
Already have an account?
Loading Related Reports