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 (Research)

** ********* *** ******* *** ********** in *** ******** ****** ******, **** entrants********* ************* **** ****** ********* ** ***** super *** **** *******.

IPVM Image

**** ** * *********** ****** *** higher ****** ********* ********,****** ****, ********* ****, *** **** ******* * ******* for ***** ******** *********.

** *** *** ***** ******** **-******** cameras ******* ** **********, ** **** compared ***** *** ************* ****** ******** categories:

  • ***** ***** ********** ****** *******, *******, rain, *******, ******, *** ****.
  • **** ****** **. ****** ***** ****** detection
  • ********* *********/*** ******
  • ************* *** **** ******** *** ****
  • *** ******* ******** *** ******* *** cloud *******

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

***** ********** ** ********** ********* ****** see ***** ****** ********* ********, ********* ********, Axis, *****, *****, ******, *********, *** Uniview, ** **** ** **** ****** AI ************ *********+************ **********.

Performance ********

***** ** *** *****, ** **** these ******* ** **** *****:

IPVM Image

  • *** ********* – ****:****** *******, **** (***** **** ***** people *********) ******* ***** ****** ** common ******* (*******, *******, *******, ****), some ** *** ** ***** ********* other ********* ******. *******, **** ****** detection ******* ****** ******* ********.
  • ******** – ****/**** **:*** **** *** **** **** ****** Detection **** ******* ***** ****** ** sources ******* ** ****. *******, **** suffers **** * ****, ***-************ * minute ******** ******* ********** ***** *** miss ***** ********, *** ** ***** in ****. ** * **** *******, this ** *********** **** ** ** issue ** ********** *** ***** ****** to ** ********, *** *** ** a ******* ** ****** ******.
  • ****** ********** – ****, ****, **:****, **** *****, *** ** **** 3 ** *********** *** *****, ******** to ***** ***** ****** ** *** common ******* ***** *****. *******, **** alarmed ********** ****** ***** **** *** Arlo's ********* ******** *** **** ** all ***** ******* ******. ************, ** frequently ******** ******* ** ****** ** its ***** ** ****. *******, **** and **** **** *** ******* ********* the ******* ******* ******* *** *****, though ******* * ******* ** **.
  • ***** ********* – *****:***** *** *** ***** ********* ** the ****, ******** ** *** ******* listed *****, ******** ** *** *** miss *** ******* ******* ******* *** scene **** ******. ******** *********** ********* the ******** ** ***** ******, *** also ******** ** ****** ********* ** subjects ** ***** *****/****** ******* *****, not **** ** *** ***** *****. [Note: ***** **** **** ****** *** vehicle ********* ********* **** ** ******** in ***** ******* ** *** ***** half ** ****.]

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

** *** ******** **** ********, ******* true ** ********* **** ****** ********* required ********* ******* **** ** **** and **** **, **** ******* *****. Now, ** ****, ** ********* *** included *** **** ** *** **** cameras **** **** *** ** **** no ********* **********. ** ****** **** segment ** ******** ** ****, **** other ******** **** ** ***** ******** to ******* ******* ** ****. ****** priced ********** **** **** ********** ******* selling ******* ******* ***** *** ********* 1/10th ** */*** *** **** ** their ***, **** ** ********* ****.

Results ********

*** ***** ***** ***** * ******* of *** *************** ** ****** *******/*********:

IPVM Image

** **** ** ***** ******* ** more ****** *****.

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

** * ***** ** ******* ***** of *******, ** **** ***** **** Arlo, ****, ****, *** **** ******** most ***** ****** ** *** **** scenes.

*** *******, *** **** *********** ******** alerts **** ******* ******* *** *** shadows *******, ***** ********** ********* **** many ********** ********* ** *** *****.

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

***** ****** **** ******** ***** ****** on *******, **** ** *** *** shown *****:

** **** ** **** ****** ****** issues **** ** ******* **** ** subjects ******* ******* ******** *****:

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

** *** *****, **** *** *********** issues ****** **** ******, ******** ****** constantly ****** ***** ****. ** ********, Arlo, **** ****** *********, *** ***** did *** ***** ** **** ** all.

Ring-Severe-Rain-False-Alerts

*******, **** **** ****'* ****** ****** detection ********* **********, ******* ** ****.

** **** ****** ** **** **** and ** ** ********* ******* **** since *** ******* *** ****** ****, but **** ** **** **** ******* models ****** *********.

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

** *** ** **** ****** *, animals ****** ******* *** ***** ********** triggered ***** ****** ********* ******, *.*. the *** *****:

*** **** **** ******** **** ***** dogs, ***** ****:

***** **'* ****** ****** ** **** surveillance, ***** ********** ** **** ** likely ** ** ** ***** *** most *****.

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

****** ******, *** ******* ****** ********** alerted ** ****** ***** ****** ** the *****, ********* *******, *******, *** simply ****** ******* ******** ******* ***** changes:

Ezviz-Alerts-On-Passing-Clouds

************, ***** *** *** **** ***** tested **** ********** ******* ** ******* at *****. ****** *** *** **** this *****.

Ezviz-Alerts-On-Insects-At-Night

******* *********** **** ****** ********** ***** alerts, *** *********** ******* *********, ** well, **** *** ****** ******* ** capture **** **** ****** ******** ** close *****.

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

***** ****'* ****** ********* *********** ** strong, **** ** ***** ****** ****** our *****, ***** ******* ****** ********* triggered ********** ** ******* *******, **** as *******, *******, *** **** (********* below) ***** ****** ********* *******.

Nest-Motion-Frequent-False-Alerts

******* ** ****, ***** ******** ****** for **** **** **** ****** (**** as ******** ** *******) *** ****** to ********** * ************ ****** ** nuisance ******.

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

**** *** ******* ********** ******** ******* subjects, ***** *** ** **** (***** below).

yi running

**** **** ***** ***** ******* ** these ********, ** **** ******* ** practically *** ***** ****** ** *** scene, **** ******** ***** ******.

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

****, ****, ****, *** **** *** failed ** ****** ******** ******* ******* the *****, **** ** ***** *****, on ******** ********.

**** ** ***** *****, **** *** subject ******* **** ~***' ******** **** the *******. ** ********, ******* *** not **** ** ****** ******** ** normal ******* ******, *** ** *****.

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

**** **** ****'* ****** ********* ********* on *** *******, *** ***** ********* person ****** **** ***** **** ***** events.

*** **** *** ****** ******** ****** across * ***** ***** ** ****, as ****. *******, ***** ***** ****** market (****** ****/***** ********** ***), ******* subjects *** **** ****** ** **** are **** ****** ** ** ** issue **** ***** ************* ******** ******* models.

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

** *** ******* ******, **** *** Nest **** ***** ** *** ***** for ******** ********** ***** ** ******** the ****. ***** *** **** ******* processing ******* ** *** ******.

**** **** *** **** **** ***** that ********** *** **** ****** **** using ****** *********, ****** ** *** little ********** ** **** ******* ***** and ************ ** *** ******* ******, with *** ****** *-* ******* ** notify ***** ** *********.

**** **** *** ********* ****** ******* cloud ************ *** ******, **** **** sending **** ************* **** *** ***** network.

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

********/****** ************* ****** ******* ******* ******, but ** ******* ** * **** (if **** ****** ** *** *******) and *********** ******** ** ****, *****, and ****.

IPVM Image

**** ******** **** *****, *** ** sensitivity, **** **** ******* *** ********* people **. ******* ******, *** ** specific ***********.

IPVM Image

**** ***** *** ******** ** *** cameras ****** ****, ***** **** **** rectangles, ****** ** ********* ** *** to *** ****** ****** ******* *********** undesired *****.

IPVM Image

** ********, **** *** **** *** eight ***** ********:

IPVM Image

***** ***** **** * ****-***** **** configuration:

IPVM Image

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

****** ****** ******* ***** ******* **** setup, **** *** ** **** ****** on/off ******** ** ****** ****** *********, which ********** ************, *** *** *** be ** ******** ** ********* ***** are *** ****** ** ******** ***** should ** *******.

IPVM Image

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

*** **** *** ** **** ******* both ******* ******** ******* ***** ****** frequent ****** ****** ** *** ***** and ****** ** *******. ** *** Wyze ***, **** ******** ** ***** at * *******, ***** ** *** timeline *****, *** ****** ** *******.

IPVM Image

*** ** ****** ******** ** * longer ** ****** ********, *** *** be ********** ** * ******* ** 3 *******:

IPVM Image

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

** ******* ******, *****, ****, *** Yi ** *** ******* * ******* subscription ** ***** ** ******* ********/***** alerts ***/** **** ***** *****.

****, ****, *** **** **, ******* from ~$* *** *** ***** *** camera *** **** ***** ******* $*/***** per ****** *** **** *****. ******* is ******** *****:

  • **** *****:**** ****** ********* ********* ******* ****** *****************, ******** ** $*.** *** ******, including * **** *******, *** ** to $**.** *** ** **** *******. Arlo ******* *** **** ******* ***** at $*** *** *** ************ ****** $*** *** ******** **** ******* ****.
  • *********/***** *******:***** ****** ********* **** *** ******* a ******* ************. ***** **** *** require * ************ ** ******* ******, with ********* ****** ** ***** ** card. ******** ***** ******* ***** (***************) ***** ** $*.** *** ***** per ******. *********'* ***** ******* ****** retails *** $***, ************* **** **** Arlo, ****, *** ****.
  • **** *** / **** ** ****** Detection:**** **** *** ****** *** ******* fees *** ********* ** ****** ** video. ***** ****** ****** ********* *** the ****** ***. *** **** *** V2 ***** ***$** *** ******** **** ****.
  • **** ****** ******:****'* ****** ********* ********* ** *** require * ************ ** ******* ******. However, ***** ***** ******** ** *********** without ***** ***** ************, ******* **** $*/***** *** ****** (5 ****) ******* $**/***** *** ** days. *** **** ******* ** ** available******** **** **** *** $***.
  • **** ******** ****** *********:****'* "******** ****** *********" **** *** require * ************ *** ****** ******. However, **** ****, * ************ ** needed ** **** ******** *****, ******** at $* *** *****.**** ********* ********** *** $*** *** *** ** purchased **** *** **** *******.
  • ** **** ****** *:**** ****, ** **** *** ******* a ******* ************ *** ****** ********* or *** ***** ******. *** ** Home ****** * ***** ***$** ******** **** **, *** *** ** ***** *** $40-45 *********.

Country ** ******

*** ********* **** ***** **** ******'* country ** ******:

  • **** *** *: **** ** *******
  • ***** *******: **** ** *****
  • **** ******* **: **** ** *****
  • **** ********* ***: **** ** *****
  • **** *** **: **** ** *****
  • ** **** ****** *: **** ** China

[****: **** ******** *** ********** ********* in ****, *** ******* ** **** to ******* ******* ** **/******** ********** and *********** *** ********** *******.]]

Comments (9)

Wyze just pushed a firmware update this week, and I haven't caught any false positives on person detection since. Was getting cars pretty regularly before from a camera that was catching cars moving along the cameras horizon.

Haven't done any digging yet to see what it has been missing.

Interesting report, thanks.

Agree
Disagree
Informative: 3
Unhelpful
Funny

I want:

1. wireless 2. individual settings per camera to turn on/off vehicle, person, package, animal, general motion detection 3. pre-alarm recording so I can see what was happening before detection. 4. at least 8 point activity detection zone setting tool. 5. truly ignore general motion if I do not set to detect (my Nest will still record and alert on motion including when there is just a pixel change due to lighting while my Arlo alarms when the leaves on the bushes move, even when I had motion detection turned off).

By the way, I do not have the same experience with Arlo on detection notification. It can take 2-3 minutes sometimes before I get an alert; 30 seconds to a minute best case.

Agree
Disagree
Informative: 1
Unhelpful
Funny

By the way, I do not have the same experience with Arlo on detection notification. It can take 2-3 minutes sometimes before I get an alert; 30 seconds to a minute best case.

Wow, really? I have an Arlo Go running on Verizon and I get alerts in a few seconds to maybe 10-15 at most. I put this camera up when we had a weeklong power outage so I basically just had some sort of view while away and it's done well for a couple years for showing me general activity.

Agree
Disagree
Informative
Unhelpful
Funny

I would like to see a test of Dahua`s IMOU too.
Imou | Enjoy Smart Life

Agree
Disagree
Informative
Unhelpful
Funny

We would consider testing the Dahua/Imou AI camera, but it doesn’t appear to be shipping in the US yet.

Agree
Disagree
Informative
Unhelpful
Funny
Agree
Disagree
Informative: 1
Unhelpful
Funny

John, thanks! We've seen the SimCam before, the marketing claims and price are impressive but I don't see any traction or a significant push from them in the commercial market. There's so many analytics to test that we are trying to focus on bigger players or those focused more on commercial, which is our focus.

We're definitely going to keep an eye out though for SimCam to see how they evolve.

Agree
Disagree
Informative
Unhelpful
Funny

Good range of models on test and matches what I’ve seen at home apart from a disappointing experience with the nest

i did have the reolink battery cameras to start but they were very temperamental.

i then got the nest outdoor its setting on a shelf just now very disappointed with the picture and the false positives from a tree in the back garden. I’ll maybe plug it back in and upgrade the software to see if it had improved

The Imou outdoor bullet pro has become available here so just ordered one to see how I get on with it.

The wyze cams for me have been the most fun I’ve got two static and one 360 model for keeping and eye on the pets when away from home during day. I get occasional false alarms from the cats moving about but the person detection is coming on leaps and bounds. They are not sold here (Scotland,UK) so had to buy them on amazon.com. I’m keeping my eyes peeled for their proposed outdoor model.

Agree
Disagree
Informative
Unhelpful
Funny

I actually have a Yi cam sitting at my desk and I prefer it to alert me whenever there is any movement. But that is exactly what I want it to do and the tracking works pretty well.

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