Consumer IP Camera Analytics / AI Shootout - Arlo, Google / Nest, Amazon / Ring, Hikvision / Ezviz, Wyze Cam, Yi Home

By Ethan Ace, Published Sep 26, 2019, 10:52am EDT

AI analytics are hitting the mainstream in the consumer camera market, with entrants Wyze and Yi Home releasing free people detection on their super low cost cameras.

IPVM Image

This is a significant threat for higher priced offerings like Arlo, Google Nest, and Amazon Ring, who have charged a premium for their advanced analytics.

To see how these newcomer AI-equipped cameras compare to incumbents, we have compared these six manufacturers across multiple categories:

  • False alert resistance versus foliage, animals, rain, shadows, lights, and more.
  • Fast moving vs. normal speed object detection
  • Detection distances/PPF levels
  • Configuration and zone creation for each
  • How pricing compares for cameras and cloud service

Cameras included:

Those interested in commercial offerings should see our IP Camera Analytics Shootout, comparing Avigilon, Axis, Bosch, Dahua, Hanwha, Hikvision, and Uniview, as well as more recent AI cameras Dahua Analytics+ and Hikvision DeepinView.

Performance ********

***** ** *** *****, we **** ***** ******* in **** *****:

IPVM Image

  • *** ********* – ****:****** *******, **** (***** Nest ***** ****** *********) avoided ***** ****** ** common ******* (*******, *******, shadows, ****), **** ** all ** ***** ********* other ********* ******. *******, Nest ****** ********* ******* missed ******* ********.
  • ******** – ****/**** **:*** **** *** **** Xnor ****** ********* **** avoided ***** ****** ** sources ******* ** ****. However, **** ******* **** a ****, ***-************ * minute ******** ******* ********** which *** **** ***** subjects, *** ** ***** in ****. ** * home *******, **** ** potentially **** ** ** issue ** ********** *** still ****** ** ** captured, *** *** ** a ******* ** ****** scenes.
  • ****** ********** – ****, Ring, **:****, **** *****, *** Yi **** * ** performance *** *****, ******** to ***** ***** ****** on *** ****** ******* noted *****. *******, **** alarmed ********** ****** ***** rain *** ****'* ********* distance *** **** ** all ***** ******* ******. Additionally, ** ********** ******** animals ** ****** ** its ***** ** ****. Finally, **** *** **** also *** ******* ********* the ******* ******* ******* the *****, ****** ******* a ******* ** **.
  • ***** ********* – *****:***** *** *** ***** performer ** *** ****, alerting ** *** ******* listed *****, ******** ** did *** **** *** subject ******* ******* *** scene **** ******. ******** sensitivity ********* *** ******** of ***** ******, *** also ******** ** ****** detection ** ******** ** close *****/****** ******* *****, not **** ** *** other *****. [****: ***** says **** ****** *** vehicle ********* ********* **** be ******** ** ***** America ** *** ***** half ** ****.]

AI **** ******** / ****** *** ****, ****

** *** ******** **** shootout, ******* **** ** analytics **** ****** ********* required ********* ******* **** as **** *** **** IQ, **** ******* *****. Now, ** ****, ** analytics *** ******** *** free ** *** **** cameras **** **** *** Yi **** ** ********* commitment. ** ****** **** segment ** ******** ** grow, **** ***** ******** such ** ***** ******** to ******* ******* ** 2020. ****** ****** ********** will **** ********** ******* selling ******* ******* ***** are ********* */**** ** 1/4th *** **** ** their ***, **** ** recurring ****.

Results ********

*** ***** ***** ***** a ******* ** *** differentiators ** ****** *******/*********:

IPVM Image

** **** ** ***** factors ** **** ****** below.

Arlo, ****, ****, ****: ***** ***** ***** **********

** * ***** ** several ***** ** *******, we **** ***** **** Arlo, ****, ****, *** Wyze ******** **** ***** alerts ** *** **** scenes.

*** *******, *** **** consistenly ******** ****** **** blowing ******* *** *** shadows *******, ***** ********** triggered **** **** ********** analytics ** *** *****.

Arlo,-Nest,-Ring--No-Alerts-On-Blowing-Foliage

***** ****** **** ******** false ****** ** *******, such ** *** *** shown *****:

** **** ** **** common ****** ****** **** as ******* **** ** subjects ******* ******* ******** rooms:

Ring **** ***** ***** ******

** *** *****, **** had *********** ****** ****** rain ******, ******** ****** constantly ****** ***** ****. By ********, ****, **** Person *********, *** ***** did *** ***** ** rain ** ***.

Ring-Severe-Rain-False-Alerts

*******, **** **** ****'* simple ****** ********* ********* frequently, ******* ** ****.

** **** ****** ** test **** *** ** AI ********* ******* **** since *** ******* *** indoor ****, *** **** to **** **** ******* models ****** *********.

Yi **** ****** *: ******** ***** ****** ** *******

** *** ** **** Camera *, ******* ****** through *** ***** ********** triggered ***** ****** ********* alerts, *.*. *** *** below:

*** **** **** ******** with ***** ****, ***** here:

***** **'* ****** ****** of **** ************, ***** detections ** **** ** likely ** ** ** issue *** **** *****.

Hikvision *****: ******** ***** ****** **** *** ****** *******

****** ******, *** ******* camera ********** ******* ** nearly ***** ****** ** the *****, ********* *******, shadows, *** ****** ****** passing ******** ******* ***** changes:

Ezviz-Alerts-On-Passing-Clouds

************, ***** *** *** only ***** ****** **** frequently ******* ** ******* at *****. ****** *** not **** **** *****.

Ezviz-Alerts-On-Insects-At-Night

******* *********** **** ****** eliminated ***** ******, *** drastically ******* *********, ** well, **** *** ****** failing ** ******* **** slow ****** ******** ** close *****.

Nest ****** **** *********** **. **** ****** *********

***** ****'* ****** ********* performance ** ******, **** no ***** ****** ****** our *****, ***** ******* motion ********* ********* ********** on ******* *******, **** as *******, *******, *** rain (********* *****) ***** person ********* *******.

Nest-Motion-Frequent-False-Alerts

******* ** ****, ***** desiring ****** *** **** than **** ****** (**** as ******** ** *******) are ****** ** ********** a ************ ****** ** nuisance ******.

Running ******** ********** ********: ** ****, *****

**** *** ******* ********** detected ******* ********, ***** and ** **** (***** below).

yi running

**** **** ***** ***** alerted ** ***** ********, it **** ******* ** practically *** ***** ****** in *** *****, **** numerous ***** ******.

Running ******** *** ********: ****, ****, ****, ****

****, ****, ****, *** Wyze *** ****** ** detect ******** ******* ******* the *****, **** ** close *****, ** ******** attempts.

**** ** ***** *****, with *** ******* ******* from ~***' ******** **** the *******. ** ********, cameras *** *** **** to ****** ******** ** normal ******* ******, *** or *****.

Arlo,-Nest,-Ring-Miss-Running-Subjects

**** **** ****'* ****** detection ********* ** *** subject, *** ***** ********* person ****** **** ***** miss ***** ******.

*** **** *** ****** subjects ****** ****** * wider ***** ** ****, as ****. *******, ***** their ****** ****** (****** home/small ********** ***), ******* subjects *** **** ****** of **** *** **** likely ** ** ** issue **** ***** ************* offering ******* ******.

Cloud ****** ***** **********

** *** ******* ******, Arlo *** **** **** video ** *** ***** for ******** ********** ***** to ******** *** ****. Ezviz *** **** ******* processing ******* ** *** camera.

**** **** *** **** warn ***** **** ********** may **** ****** **** using ****** *********, ****** we *** ****** ********** in **** ******* ***** and ************ ** *** cameras ******, **** *** taking *-* ******* ** notify ***** ** *********.

**** **** *** ********* tested ******* ***** ************ for ******, **** **** sending **** ************* **** the ***** *******.

Motion/Analytic ************* ***********

********/****** ************* ****** ******* cameras ******, *** ** limited ** * **** (if **** ****** ** not *******) *** *********** settings ** ****, *****, and ****.

IPVM Image

**** ******** **** *****, but ** ***********, **** only ******* *** ********* people **. ******* ******, but ** ******** ***********.

IPVM Image

**** ***** *** ******** in *** ******* ****** Arlo, ***** **** **** rectangles, ****** ** ********* to *** ** *** shaped ****** ******* *********** undesired *****.

IPVM Image

** ********, **** *** Ring *** ***** ***** polygons:

IPVM Image

***** ***** **** * grid-style **** *************:

IPVM Image

Wyze *** ** ** **** *****, ****** **/***

****** ****** ******* ***** require **** *****, **** and ** **** ****** on/off ******** ** ****** person *********, ***** ********** installation, *** *** *** be ** ******** ** detection ***** *** *** shaped ** ******** ***** should ** *******.

IPVM Image

Wyze / **: ******** ******* ***** ****** ******

*** **** *** ** Home ******* **** ******* cooldown ******* ***** ****** frequent ****** ****** ** our ***** *** ****** be *******. ** *** Wyze ***, **** ******** is ***** ** * minutes, ***** ** *** timeline *****, *** ****** be *******.

IPVM Image

*** ** ****** ******** to * ****** ** minute ********, *** *** be ********** ** * minimum ** * *******:

IPVM Image

Camera / ******* *******

** ******* ******, *****, Wyze, *** ** ** not ******* * ******* subscription ** ***** ** receive ********/***** ****** ***/** view ***** *****.

****, ****, *** **** do, ******* **** ~$* USD *** ***** *** camera *** **** ***** through $*/***** *** ****** for **** *****. ******* is ******** *****:

  • **** *****:**** ****** ********* ********* require ****** *****************, ******** ** $*.** per ******, ********* * days *******, *** ** to $**.** *** ** days *******. **** ******* and **** ******* ***** at $*** *** *** original**** ****** $*** *** ******** **** ******* ****.
  • *********/***** *******:***** ****** ********* **** not ******* * ******* subscription. ***** **** *** require * ************ ** receive ******, **** ********* direct ** ***** ** card. ******** ***** ******* plans (***************) ***** ** $*.** per ***** *** ******. Hikvision's ***** ******* ****** retails *** $***, ************* less **** ****, ****, and ****.
  • **** *** / **** AI ****** *********:**** **** *** ****** any ******* **** *** analytics ** ****** ** video. ***** ****** ****** analytics *** *** ****** app. *** **** *** V2 ***** ***$** *** ******** **** Wyze.
  • **** ****** ******:****'* ****** ********* ********* do *** ******* * subscription ** ******* ******. However, ***** ***** ******** is *********** ******* ***** ***** ************, ******* **** $*/***** per ****** (* ****) through $**/***** *** ** days. *** **** ******* IQ ** ***************** **** **** *** $349.
  • **** ******** ****** *********:****'* "******** ****** *********" does *** ******* * subscription *** ****** ******. However, **** ****, * subscription ** ****** ** view ******** *****, ******** at $* *** *****.**** ********* ********** *** $*** *** can ** ********* **** the **** *******.
  • ** **** ****** *:**** ****, ** **** not ******* * ******* subscription *** ****** ********* or *** ***** ******. The ** **** ****** 3 ***** ***$** ******** **** **, *** *** ** found *** $**-** *********.

Country ** ******

*** ********* **** ***** each ******'* ******* ** origin:

  • **** *** *: **** ** *******
  • ***** *******: **** ** *****
  • **** ******* **: **** ** *****
  • **** ********* ***: **** ** *****
  • **** *** **: **** in *****
  • ** **** ****** *: Made ** *****

[****: **** ******** *** originally ********* ** ****, but ******* ** **** to ******* ******* ** AI/analytic ********** *** *********** our ********** *******.]]

Comments (9)

**** **** ****** * firmware ****** **** ****, and * *****'* ****** any ***** ********* ** person ********* *****. *** getting **** ****** ********* before **** * ****** that *** ******** **** moving ***** *** ******* horizon.

*****'* **** *** ******* yet ** *** **** it *** **** *******.

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

* ****:

*. ******** *. ********** settings *** ****** ** turn **/*** *******, ******, package, ******, ******* ****** detection *. ***-***** ********* so * *** *** what *** ********* ****** detection. *. ** ***** 8 ***** ******** ********* zone ******* ****. *. truly ****** ******* ****** if * ** *** set ** ****** (** Nest **** ***** ****** and ***** ** ****** including **** ***** ** just * ***** ****** due ** ******** ***** my **** ****** **** the ****** ** *** bushes ****, **** **** I *** ****** ********* turned ***).

** *** ***, * do *** **** *** same ********** **** **** on ********* ************. ** can **** *-* ******* sometimes ****** * *** an *****; ** ******* to * ****** **** case.

** *** ***, * do *** **** *** same ********** **** **** on ********* ************. ** can **** *-* ******* sometimes ****** * *** an *****; ** ******* to * ****** **** case.

***, ******? * **** an **** ** ******* on ******* *** * get ****** ** * few ******* ** ***** 10-15 ** ****. * put **** ****** ** when ** *** * weeklong ***** ****** ** I ********* **** *** some **** ** **** while **** *** **'* done **** *** * couple ***** *** ******* me ******* ********.

* ***** **** ** see * **** ** Dahua`s **** ***.
**** | ***** ***** Life

** ***** ******** ******* the *****/**** ** ******, but ** *****’* ****** to ** ******** ** the ** ***.

****, ******! **'** **** the ****** ******, *** marketing ****** *** ***** are ********** *** * don't *** *** ******** or * *********** **** from **** ** *** commercial ******. *****'* ** many ********* ** **** that ** *** ****** to ***** ** ****** players ** ***** ******* more ** **********, ***** is *** *****.

**'** ********** ***** ** keep ** *** *** though *** ****** ** see *** **** ******.

**** ***** ** ****** on **** *** ******* what *’** **** ** home ***** **** * disappointing ********** **** *** nest

* *** **** *** reolink ******* ******* ** start *** **** **** very *************.

* **** *** *** nest ******* *** ******* on * ***** **** now **** ************ **** the ******* *** *** false ********* **** * tree ** *** **** garden. *’** ***** **** it **** ** *** upgrade *** ******** ** see ** ** *** improved

*** **** ******* ****** pro *** ****** ********* here ** **** ******* one ** *** *** I *** ** **** it.

*** **** **** *** me **** **** *** most *** *’** *** two ****** *** *** 360 ***** *** ******* and *** ** *** pets **** **** **** home ****** ***. * get ********** ***** ****** from *** **** ****** about *** *** ****** detection ** ****** ** leaps *** ******. **** are *** **** **** (Scotland,UK) ** *** ** buy **** ** ******.***. I’m ******* ** **** peeled *** ***** ******** outdoor *****.

* ******** **** * Yi *** ******* ** my **** *** * prefer ** ** ***** me ******** ***** ** any ********. *** **** is ******* **** * want ** ** ** and *** ******** ***** pretty ****.

Read this IPVM report for free.

This article is part of IPVM's 6,653 reports, 896 tests and is only available to members. To get a one-time preview of our work, enter your work email to access the full article.

Already a member? Login here | Join now
Loading Related Reports