Carnegie Mellon AI Startup Zensors Profile

By Sean Patton, Published Jun 11, 2019, 11:40am EDT

Zensors is a startup formed by Carnegie Mellon graduates from a Carnegie Mellon research project, offering customized models per camera that they claim delivers significantly higher accuracy.

We spoke with Zensors to better understand their approach, how they plan to scale their system and their go-to-market.

In this note examine:

  • Who is Zensors
  • What Customers is Zensors Targeting
  • How Much Does It Cost
  • What is Zensors' Go-To-Market
  • What Type of Analytics
  • How Do They Generate Custom Datasets
  • What VMS Integrations

Zensors ********

******* ** ****, ******* was ********** * ******** project ** **-************* ****************** ****. ******* ******* ** employees ** ********** *** San *********, ********* ******* development ******, *** * 20 ****** *********** **** in ***** *** ******* labeling. ******** ** ** Assistant ********* ** *****-******** Interaction ** ******** ****** University, ********* ***** ****** Interfaces *****, ***** **** earned * ******* ****** in *****-******** *********** ** Carnegie ******, ** **** as ********** ******* *********** and ** ** ******** at ********.

******* ******* $*.* ******* in *******. **** *** not ******** *********, *** they *** ******* *************, *** *** * partnership **** ***, *** 2 ****** *********** ****, ******* ** ***** city ********** ***********.

***** ********* **** ******* Into *******

******* ** ******** ***** AI-based ***** ********* **** uses ******-********* ****** ******** that *** ********* ******* to ****** ****** *** vehicles, *** ***** **** 500 ****** ** ******* detection *** *********.

******* **** **** ***** their ****** ** *** customer's ******** ******* ** generate ************ ****** ***** on ****-***** ***** (*.*. real-time **** ********, ******* tracking, **** *********). ****, they ***** **** *** developing ****** ******** ** analytics, ***** ** ****** needs ****** ********* (*.*. Smart **** *******, ************** Terminals).

**** *-****** ********* ***** shows ******* ******-******** *** queue ********** ** ****:

One ***** *** ******

******* **** ******** ****** are ******* *** ********* per ******. **** ** atypical **** **** ***** analytics, **** **-***** ******, which *** ******* ** public ** ****** ******* sets ** ****** ** detect *** ******** ******* (for **** ******** *******, read ******** ******** ******** *** Video ************). ******* ****** **** contributes ** ****** **** are **-**% **** ******** and ******** ******** **** poor ********, ***** ****, and ****** *********. **** noted ***** ** * baseline "********" ****** ******* model *** ***** ******** per ****** ** *** a "**********", *** ******* it **** *** ******* as **** ********* ** training *** ****** ****** network.

******* **** ******** *** training *** ******* ***** publicly ********* ***** *******, customer ****** ***** *** images ****** **** ******* installs ** ***** ******** sites.

****** ********* ** ****** Questions

******* ****** * *** dashboard *** ***** *******, searching *** ****** ************. Their ****** ** ****** on ***, *** ***** it *** ** ********* on-premise, ******* **** **** have **** ******** ******.

******* **** ***** ********* is ******** ** ***** high *****, **** ** digest ****** ** ***********, by ******** *** ***** analytic ****** ** ****. To ****** *** ******, a **** **** * question (*.*. "*** **** cars *** ** ****?", "how **** ****** *** here?"), *** *** ********* specific ***** ** *** video ** ******:

**** ** ******* ** functionality ******* *******'* ******* ******** ********** (IPVM ****). *** ******** *** be ******** **** **** Zensors ***** "********* *****" Reports, ***** ******* ********** summaries ** *** ****** in ******* ******* *******:

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

******* **** **** ***** not ** ******* **** recognition ****** ** ****** classifications, *** ******* *** abuse ********. **** **** said ***** ********* *** not ******* ** ******* anomaly ********* (*.*. ******* behavior, ******* ** * library) ** *******, *** instead *** ********** ****** training ** ******** ******* behaviors.

Targeting ***** ******, *********, ***** **********

******* **** **** *** targeting ***** **** ********, transportation **** (*.*. ********, train *********), *** ***** facilities **** ***** ******* of ****** *** ******** (e.g. **** ******** *******, urban ******* *******). * differentiator ****** ****** ** the ******* ** ******* vehicle ******** ********, ***** they **** ****** **** to ********** ****** ********, and ******** *** ***** of ********. **** **** they **** ********* **** their ********* *** ***** parking ************, *** ****** and ******* ********.

******* ****** ** ******* city **** ******* * IP ****** ****** **** an ********** ******** *** connection, ********* **** ** city *******, ***** ***** $1,500 - $*,***. **** said **** *** ***** by * ***** ******** in **********, **** ***-***-***** components.

Per ****** *** ***** *******

******* **** **** ******* pricing *** ******** ******** models (*.*. ****** *** vehicle ********) ** $** per ****** *** *****. Subscriptions *** *** *****, so * ****** ****** counting ****** *** ******** would ** $** *** month. ******* **** *** volume, ********** ********-***** ****** cost $** *** ***** per ****** *** *****. They **** **** ***** are ******* ***** *********, which ****** ** *** amount ** ********** ********, based ** ***** ******** and ******** ***** **********.

3rd ***** ****** *******, ********* ***** *******

******* **** **** ******* 3rd ***** ******* ***** multiple ********* (*****, ****, FTP ********), *** ********** to *******. ******** *** many ***** *********, ******* said **** ** *** require **** ***, *** noted *** ********** (*.*. people ***** ********), * frame ***** ** ******* is ******. ******* ******** and ******** ***** ********* require ****** ***** *****.

** ******* ******* ** Zensors *******, **** ******* an **-******* ******* **** 6 ****** ** ****** contracts.

Direct ***** *********, **** ** *** ****

******* ***** ****** ** end ***** *** ***** they *** ******** ** build ** ********** *******. They **** **** ************ an *** ***** *** integrators ** ***** ******* sales ****, *** **** are ********* ****** * sales ***********/******** *******,** ******** ** **** development ********* *** ***********.

Only *** *********** - **********

******* ***** ***** **** VMS *********** ** ************** ************ **** **** ***** be ********** **** ************ as **** ** ***** development *******. ************, **** said ****** ************ **** being ********** ** ******** connect *** **** ********* direct **** *** ****** to ****** *******.

Potential **********, ** ** ****** ***** ****

***** **** ******** ****** is * ****** ** AI ********, ** ******* that *** ********** ** leading ****. *** ********* we *** ** **** it ** *** ***** what ******** ****** **** are ** (** **** a ****** ********? ***********? transportation?). **** ******** ***** may **** ***** ** a **** **** **** who **** ** ****** work *** **** (******* example****** ***** **** ****** Analytics ****). *******, *********** ********, they *** ******* **** a **** ****-*** *********** specific ******** **** ** easier ** ********** *** to ****.

Comments (1)

John Honovich

IPVM | 07/31/19 06:29pm

******:********** ************* ******* ******** with ********** ************ ******* Zensors ** ******* **** Accurate **** ***** *** TSA *****

**** ***** ***** *******, an ***** ************* ** in ******:

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