YCombinator AI Startup Visual One Tested

By Rob Kilpatrick, Published Apr 02, 2020, 11:14am EDT

Startup Visual One, backed by Silicon Valley's powerful Y Combinator, aims to be "Your 24/7 Watchman" with advanced analytics and object recognition for people, children, pets, wildlife, and more, working with their own OEMed cameras and compatible with Nest Cams.

IPVM Image

IPVM tested Visual One analytics with our Nest cam examining:

  • How does this object detection work in real world scenes with animals and common objects?
  • Does it ignore common indoor and outdoor false alert sources?
  • Does it misclassify people or animals?
  • Can it detect people crawling, running, and obstructed?
  • How hard is it to set up an event?
  • What PPF levels are required for detection?

We also obtained feedback while testing and prior to publishing from founder Mohammad Rafiee.

*******

*******, ** ***** **** has **** ****** **** term ********* ** ** delivered ******** ********* ***** accurately. *******, ** ** a *******, **** *** early ***** *** **** need ** **** **** they *** ***** *** the ****** *******, ******* it *** ****** ** to * ****** ****** (whether ****, *** ** enterprise).

***** ** *** *****, Visual *** *** **** to ****** **** ******* in **** ******** ********* other ********* ** ***, such ** ******* ** top ** ********* *** people **** ******** *****, without ***** ****** ** common ******* **** ******* foliage, *******, ******, ***, which *** ** ****** in **** ** ******* DIY ************.

*******, **** ********* **** support**** ******* ****** ***'* *** consumer/DIY-focused *******, *** *** not ********** **** ***** or **** *******, ******** their ********** ** ****** commercial ************ (***** **** do *** ****** ******* today ******. ************, **** currently **** ***** ******** notifications, **** ** *** to ****** ***** ******, and ****** *** **** stored *** ** *****.

Positives *******

****** ***'* *** ********* in *** ******* ********:

  • ******/****** **** ** ****** Accurate:******** ******/******** ****** **** as ****** **** ** car ** *** ** couch ********* ********** ** our *****.
  • ** ***** ****** ** Outdoor *******:******* ******* ** ***** alerts **** ** ******* foliage, *******, *** (*********) animals **** *******.
  • ** ***** ****** ** Indoor *******:****** ******* ** ***** alerts **** ** ***** changes, *******, *** ****** from ******** ***** ***** turned ** *** *** were *******.
  • *** *** *** ****** Recognition:**** * *** *** required *** ******** ****** detection.

Negative *********

** *** **** **** key ********* ******:

  • ********* ** ***** *********:** *** **** ** testing, ****** *** ***** not ****** ****** **** were ******* ** ****** they *** **** *** working ** ****.
  • ** *** ** ****** Past ******:**** * ****** ***'* worth ** ****** *** be ********, **** ** way ** ****** ******** day ***** **** **** say **** *** ******* on.
  • ***** ****** ***************:****** ** ****** ********, cats **** ********* ********** as **** ****** *******.
  • ******* ************ ****** ********* Frequent ***** ******:******** ************ ****** ***** alarms **** **** **** the ****** *** ***** in *** *****.
  • **** ****** *** ******** Public *******:** ***** ** ********* with **** *******, ****** the ****** ****** (******** by ****** **** *** provided ****) ** ********.

****** *** ***** ** they *** ******* **** adding ***** ****** *** longer ***** ******* ********, and *** ** *********** with ***** ****** ************* to *** ********** *************.

*******

****** *** *** ** annual ************ ** $***-*** per ******/*******. ***** **** events ******* ***** ****** has * ******* *** of $*.** *** ***** per ******, ******* ** Nest's *** **** ***** plans ($**/***** *** ***** camera, $*/***** *** **********) which ****** ** **** of ***** *******.

Event ***********

** ****** * ****** event, ***** **** *** objects, **** ** *** and ****** *****, *** select *** ******** ************ that **** **** ** be ******** **, **** as **, *****, **** to, ***. ******* ***** such ** ****** ** vehicle ********* *** **** possible.

**** **** ****** *** includes *** **** ****** type *********** **** **** analytics, **** ** *****, clock, ****** *****, ******** bat, ****, ****, ***.

Object ** *** ** / **** ** / ***** ********

********* ******/******* ** *** of, *****, ** **** to ******* *** ******** in *** *******, *** example, **** **** ********** detected ** *** ** a ****** **** *****.

IPVM Image

**** *** ****** **** also ********** ******** **** to * *** ** truck.

IPVM Image

Animals ********* *************

************, **** **** ************* as ****, *** **** was ********, ********* **** a *** ***** ** several **** ** *****.

IPVM Image

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

****** ******* ******* ** false ****** **** ******* such ** ******* ******* and ******* ** *** scene.

Common ******* ** ****** ***** ****** *******

*******, ****** ******* ** false ****** **** ******* such ** ****** ******* on *** *** ** a ****:

** ******* **** **** other *****:

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

*******, ***** **** *** key ********* ****** ** our *******. *****, ****** One ****** ****** ******* and ******** **********.

IPVM Image

****** *** **** **** plan ** ******* ***** issues ** ****** *******.

Objects ************ ***** ******** ***** ******

**** ********** ************ ******* in *** *****, ******** "Object ***********" ****** ******** while *** ****** ** still ******* ** *** scene. *****, * ******* and ******** *** ** be ******* ** ** they ***** *** ***** but ******** ****** *** sent **** **** *** still *******.

****** *** ************ ***** issues *** **** ** improve ****** ************* ****** in ****** *******.

IPVM Image

Low *** ****** *** ****** *********

****** *** ********* **** needed * *** *** person *********, ** *** testing, ***** *** **** IQ (***° ***) **** meant * ********* ******** of ~**'. **** *** level *** ***** *** compared ** **** ******* commercial ********* **** **********'* *********** ***.

IPVM Image

App ******

*********, ***** *** ** recordings ** ****** **** using **** ******* ** the ***, **** ********* included **** *************. ****** One ***** **** ********* they *** ****** ************ recording ***** ***** **** Nest **** *** *** working ** ********* **** issue.

IPVM Image

************, *** *** **** saves *** **** ****** events, *** **** ****** see *** ** *** past ******. ****** *** noted **** ******* **** alerts **** ** ***** soon.

Nest ******* ******* ****** *******

** ***** ** *** Nest ******* ** *** Visual *** *** ****** sharing ** *** **** feed ** ********, ******* anyone **** *** **** to *** **** *** view *** ******. ****** One ***** **** **** are ** *********** **** Nest ** ****** **** requirement *** ******* ** other ****** *******, ** well.

Version ****

*** ********* ******** **** used ****** *******:

  • ****** ***: ******* *.*.*

Comments (3)

**** ** **** ****, amidst *** *** *********** craziness. * **** ****** see ****. * **** hope *** ******* ****** hard ***** ******** ** the ********** ***** ***** I ***** ***** ******** would ** **** ****** valued.

**** **** ****** *** includes *** **** ****** type *********** **** **** analytics, **** ** *****, clock, ****** *****, ******** bat, ****, ****, ***

******* ******* ** *** kitchen **** * ****** ;)

******: ** ********** ****** pricing ** $*.** *** month. **** ** *** events ******* *** *******, and *** ********* ** an ****** ************ *** $300-500 *** ******/*******.

Read this IPVM report for free.

This article is part of IPVM's 6,803 reports, 913 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