Axis AI-Based Object Analytics Tested 2021

By Rob Kilpatrick, Published Jan 28, 2021, 10:13am EST (Research)

**** *** ******* ******** **** ******** analytics ** ********** ** *********** ****** *********, *** **** ** **** ** to ***** ****** ** "******* ************"?

IPVM Image

** **** ***, ** ****** *** tested *** **** ******** **** ***** Mk *** *** ****** ****** ********* on *** ******* ******** ******* *****-*, examining:

  • *** **** *********** ****** ******* **** learning *** ******* ******** ******?
  • *** ** *** ******* ****** ****** outdoor ***** ***** ******* **** *******, foliage, *** *******?
  • *** **** ** **** ****** ****** false ****** **** *******, ******, *****, etc.?
  • *** **** ********* ****** ****** *******, crawling, ** **********?
  • *** ****** *** ******** ******** **********?
  • *** *** *** **** ********* ****** people?
  • *** **** ***** ******* *******?

****: **** **** *** ******* ******* while**** ************ **** ******** ***** ** *** in **** **** *** ******* ** shortly **********, *** **** ******** ********* for ** *** **** ******** **** month, ******* ****.

*******

**** ****** ********* ** *** **** learning ******** (****) ***** ** *** was **** ****** ******* ***** *** machine ******** (****) ***** *****-* *** some ************** ********.

*** ***** ** *** ******* ***** alerts **** ****** ******* ******* *** out, ********* *******, *******, *******, ***** changes, *** ******, *** ********** ******** subjects **** *******, ********, *** **********. Vehicle *************** ** *** **** ******** model **** ********, ******** *********** ****, trucks, *****, *** *****.

*******, ** *** ******* ******** (****) models *****-*, **** ****** ********* ****** to ****** ******** ********, ************ ************* people ** ******** *** **** *****, and ******** ***** ****** ** ***** animals, *********** **** ** ******. *******, it *** *** ******** ****, ****, shadows, ******, *** ** *******.

**** ********* ******** ~** *** *** accurate ***** *********, ******** ** *** tested ********** **** ** ******** *** or ***** ***.

****** ******** ****** ********** **** *******, Milestone, *** ** ******* ** *** tests, *** *** *** **** ** Avigilon ******* ****** ** *****.

Vs. ********, *****, ******

** ***** ** ******** *** ***** alert ***********, **** ****** ********* ** deep ******** ****** **** *** ***** Mk *** ** *********** ******* ********** ** **** *****, **** **********’* ***,***** ***,***** *********+, ********’* **, **** ****** ***** ***** **********, accurate *********, *** ******* *****/*** ************.

*******, ** ****, ***** ************* *** offer ******** ***** *** **** ***, such ** ******** ****** ******** (*****, clothing ****, ******* ****, ***.) ***** can ** ****** ** *** ******** by *****. **** ****** ********* ***** only ***** ***** **** (**/***) ** VMSes, **** ** ****** ** ****** metadata.

Axis **** ******** **. ******* ******** ******

**** ****** ********* ********* **** ******** a ******** ** ****' ******* ******* and *** ***-********** **** ******** ******* on **** * ****** **** ******* 30 ********** *** ******* ******** ******.

IPVM Image

*** *** ** ************ ** **.*/**.* firmware, **** ** ******* ***.

*** ** ***** ****** *******' *** ******-* ***, *** **** ******** ****** ******* a ****** ********* (****). **** *** declined ** ***** **** **** ** used *** *** ****.

*** ********* ***** **** *** ** found ****. **** **** **** **** **** be ********* **** **** ******** ****** in ****, ********* *, *, *** Q ****** *******.

Deep ********: ** ****** ** ****** *******, ********, ********** ** *****

** * **** ** *******, *** DLPU ****** *** *** **** ****** running ** ******** ******* *** *****.

IPVM Image

Machine ******** ****** ****** ****** ********

*******, *** **** (******* ******** ********** unit) ****** ****** ****** ******** ******* the *****, ***** ****:

IPVM Image

**** ********* **** **** *** ******** behavior *** ******* ******** ******, *** that **** *** ******* **** ********* detection ** ****** ********:

**** ** ******** ******** *** **** camera (****). ******** ** *** ** specifications ***** *** ***** ** ** based ** ** **** ******* *** upright *******. **** ** **** ********* in *** **************. *******, ** *** looking **** *** ** *** ******* this ****** *******.

Obstructed ****** ********

**** **** ******** *** ******* ******** models ********** ********* ****** **** ************* obstructed:

IPVM Image

MLPU ****** ********** ***** ** *******, ******* ** *****

** **** ********* *** **** ****** classified ****** ** ********, ***** ***** the **** ******** *** *****:

IPVM Image

*** ** **** ***** ******** **** classified ** ******, ** ****** *** cases ** *****, ***** ****:

IPVM Image

**** ************ **** **** **** ** misclassification *** ********* **** **** ***** of, ** ****:

*** **** ** ***************** ** ********* we *** ***** ** *** *** continuously ******* ** *******. **** ****** Analytics ******* ** **** ******* **** as *** **** *****/-* ** ***** on * **** ***** ****** ***** compared ** *** **** *******, *** this ** ********* **** ************* *** occur ************. ********** ** *** ****** is ******* ** ******** ***** ** some ***** ****** ***** ***** *** contour **** *********** ** ******** *********. We **** **** ***** ******* *** AOA ***** ** ******* ******** **** more **** ** **** ******. *** more ********* ****** **** ****** ************ we ********* **** ** *** **** cameras.

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

******* ****** ********* ***** ****** ** indoor ******* **** ** ***** ***** changes, ***** ****:

IPVM Image

** ******* *** ****** **** ** the *****:

IPVM Image

Deep ******** - ** ***** ****** ** ****** ******* *******

****** ******* ******* ** ***** ****** such ** ******* ******* *** *** shadows ******** **** ******* ** **** the **** *** **** ******.

IPVM Image

*** **** ******** ****** **** *** not ***** ***** ** *** ******* in *** *****.

IPVM Image

****** ** *** ********, *** ****** of * **** ******* *** ****** classified ** * ******. **** ******** only **** ** *******, ******* ***** deer ***** ** *** *** *** ~4 *****.

IPVM Image

Machine ******** - ***** ****** ** *******

** *** *****, *** ******* ******** camera ******* ********** **** ** ******, shown *****.

IPVM Image

No ***** *************** ** ***** ****/****

* ****** ****** ** ***** ****** for **** ***** *********, ***** **** and **** **** ******* ** **** Axis ******, **** ****** *** ******** accurately ********.

IPVM Image

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

** **** ******** *******, **** ****** Analytics ******** ******* ****, ********* ***, truck, ***, *** ****, ***** ******* learning ****** ******* **** *** ******* "vehicle" **************.

** *** *******, *** **** ******** camera ********* ********* ********** ******* ****, such ** *** *** *****:

IPVM Image

"*****" ************** *********** ***** ***** ****** and **** ** *** *******, ********** shown ****.

IPVM Image

******* *** **** **** ********** ** cars ** *** *******, ********* **** Axis.

IPVM Image

****** *******, ** ******** * ***** amount (were ********** ** ****, *** **** was **** ****, *********** ** ******** hundreds ** ******** ** *** *****.

IPVM Image

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

** ** ****** *** ***** ** the ****** *** *** *** **** itself **** ** *** **** *** setting **************. ** *** ******* ***** is *** * ***** ********** ** SUV´s *** ******* *** **** *** also **** ******* ********* ******. **** together **** *** **** **** ******* may **** *********** **** ********* ******, makes ** *********** *** *** *********** to **** * ********** ******** ** all *****. ** ******* ** ***** expect ***´* *** ******* ** ** classified ** **** *** ***** ** angle ** ********** *** ***** ******* just *********, **** *** ****.

~11 *** ******** *** ****** *********

****** ******* ~** *** *** ****** for ****** ********* ** *** **** and ~** *** *** ****** ** the ****.

IPVM Image

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

****** ********* ****** ********** **** *******, Milestone, *** ** ******* ** *** testing, *** *** ******** ** *****.

IPVM Image

** ********* ***** (*.*. *******, ***** below), ***** *** ****** *** ******* event ******* *** ***, ** ******** rules (*.*., **** ******, **** ********, etc.).

IPVM Image

**** **** **** ****** **** ****, e.g., ****** ******** ** ***** ********, but *** ****** ******** ** ** common **** ******* *********** ******* **** as****************.

Burnt-In ******** *****

**** ** *** ***** ****** ******** bounding *** *********** *** ********, ******. However, **** ****** ********* ****** ******** boxes ** ** "***** **" ** one ** *** ******'* *******, ** it *** ** ******* ** *** VMS.

IPVM Image

Simple *****

***** ** ******* ** ***** *********, effectively **** ********* ***** ** ****** what ******* ***** **** ** ****** (people/vehicles) *** *** **** ** ****** in. **** ****** ** **** **** as * "********" (***** *****), *** users *** **** ** ** * scenarios *** ******.

**** ******** ******** ******** **** ***** previous ********* **** ** ***** ****, swaying *******, ***** *******, ** *** ability ** *** *** ****** ** different ******* ** *** *****. *******, we ***** ** *** ******* **** these *** *** ****** *********** ** either *** **** ** **** *******.

IPVM Image

Q1615 ** *** ******* ***********: ****** *** *** ***** *** *****

*** **** ***** ** ***'* *** details **** ******, ******* ** **** cameras ** *** **** *-****** ****.

IPVM Image

***** ****, ** *********** *** ***** along **** *********** ****** **** ****** and *********.

IPVM Image

************, ****** ** ***** ** ~* lux *** **** ******, **** ****** and ***** ******* ****** **** ***** models ******, *****:

IPVM Image

*******, ** **** *** ***** (~*.* lux), *** ***** ** ***, ****** facial ******* **** ** ******** *** colors ******* ********, ***** ****** *** Hikvision ****** *** ********, ****** **** much **** *****, ******** *******.

IPVM Image

*******, *** **********, **** ***** ******* appeared ****** ** *** ******* ******, with *** ******* *** **** ***** clearly ******* *** *******.

IPVM Image

Versions ****

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

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

Comments (7)

*'* ** ********** ** **** ******* Axis ****** **** ******** ***** *******? Looking ** ***** ******* ** ******* that *** *****-*** ** * **** camera **** **** ******** *** **** a ********* **********. ** ********* ******* dome ******* ** *** *-* ** range.

Agree
Disagree
Informative
Unhelpful
Funny

* ***** * **** ** ******** read *** ******* ******* ** ******** next ****. *** ******* ******* **** :)

Agree
Disagree
Informative
Unhelpful: 1
Funny: 3

**, ** ** *** ****. **** hasn't ******* *** ****** ** ******* of ***** **** ******** ****** ** any ******* ***********.

*** ***** ** ***** $*** **** than *** ***-**** ******** *****, ** that *** **** ** **** ****. It **** ** **** *********** **** they ******* * ****** ******. ******* to *** *** ****'** ******.

Agree
Disagree
Informative
Unhelpful
Funny

*'* ******* ** **** ** ***. My ************ **** ****** ********* *** a *** ***. *'* ** *** point ***** ** * **** ******* by ***'* ********** ** * ***, I'll ** *********.

Agree
Disagree
Informative
Unhelpful
Funny

**** **-****** ********* **** *** **** testing **** *** **** ************ **** low?

Agree
Disagree
Informative
Unhelpful
Funny

** ***** ** *********** ** **** when **** **** ****** *** **** cameras **** ****. **** ****** **** more?

Agree
Disagree
Informative
Unhelpful
Funny

**** ***********,***** ***

Agree
Disagree
Informative
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