Deep Learning Tutorial For Video Surveillance

Author: Brian Karas, Published on Oct 17, 2017

Deep learning is a growing buzzword within physical security and video surveillance.

But what is 'deep learning'?

In this tutorial, we explain deep learning specifically for video surveillance covering:

  • Traditional video analytic approaches
  • Machine learning vs deep learning
  • What makes learning 'deep'?
  • How deep learning can help analytics
  • The role of training
  • Examples of training for people and guns
  • Example of training for men vs women vs old vs young
  • Filtering alarms With deep learning
  • Training data not disclosed
  • Potential Problems Across Regions
  • Hardware Requirements
  • Evaluating Deep Learning Products

**** ******** ** * ******* ******** ****** ******** ******** *** video ************.

*** **** ** '**** ********'?

** **** ********, ** ******* **** ******** ************ *** ***** surveillance ********:

  • *********** ***** ******** **********
  • ******* ******** ** **** ********
  • **** ***** ******** '****'?
  • *** **** ******** *** **** *********
  • *** **** ** ********
  • ******** ** ******** *** ****** *** ****
  • ******* ** ******** *** *** ** ***** ** *** ** young
  • ********* ****** **** **** ********
  • ******** **** *** *********
  • ********* ******** ****** *******
  • ******** ************
  • ********** **** ******** ********

[***************]

********

**** ******** *** ****** * *** ** ********* ** *** security ******** *** *** ********* ** ******* ******* ** *******-*** analytics ********. ** ** ********** ** ** *** ****** ********* of ********** ************ ** ******* ******** *** ** ***** ******* in * ******* ** ********** ** **** ********* ****** ********** and ******** ******* ****-***** ******** *** **********. *** *** ******** of **** ******, ** *** ********** **** ******** ************ ** the ******* ** ***** ************, ****** ********* *****.

Traditional ***** ********* **********

***** ** **** ******** *** ** *********, ******* ********* ****** on **** ***** *********** ** **** ******* ** ****** (*.*.: people *** * ******** ****** *****, ******** *** * **** aspect *****), *** ****** *********** ** **** ******. *** *********** step ***** ****** ***** *** ****** *** ******* ****, ** that *** ****** ***** **** ************* ****** ******* ***** *** ground (*****, ******, ***** ******). *********** **** ****** ****** **** filters, ***** ** **** ******* ***** ** ******* ** **** were *** ***** ** *** ***** *** ************ ** * given ****.

***** ********* *** ***** ****** **** *** ** *** ***** place ** *** *****, ** *** ***** ****, *** ********* motion ***** ** ******** ** ********* ** ** *** ** object ** ********. ** **** *****, ******* **** ****** ***** or ***** ********* ***** ** **** ** ********* ** ** object *** * ******, ** *******, *** *******. *******, ***** systems ********* *** *** **** * *** **** * ***** or * ********** **** * *******, ** **** *** *** doing **** ******** ** ****** ******* *** ******* ******* **** at ****** ***********.

***** ******* *** **** ***** ** ****************** **** ******* ** not **** *** ***-*** ************, * ****** ******** ** *** ground ******* ****** ******* ******** ***** ** ********** ** * vehicle ******* ** * ******, ***** ***** ****** ***** *** uniform ********:

**** ******** ** ********* *** *********** ******* ***** ** ***-********* static ******** ** ********* ******** ** ** ******* ********.

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

******* ******** *** **** ******** *** *******, *** **** ******** different ********** ** * *******. ******* ******** **** ***-********** ************ to ***** * ******** ** ********* ** ***** ** ** object, **** ******** ******* ***** ****** *** *************.

******* ******** *** ** ****** *** ** ********* * ***** walking ** *********** ********** **** ** *** ***** ** *** height ****** ** ****** **** *** *****, ***** ****** ** movement ** **** *** **** **** ** **********, ** ****** move ** * *********** ********* ******* ** ********, ** ****** have **** ***** *** ******* ******* (************ ********), *** ** forth. **** *** ********* ** **** *** *****, ** **** look *** ***** **********, *** ** ** ***** ****** ** them, ** **** ****** *** ***** ******** * ****** *******.

** **** ********, *** ******** ** *** ****, *** **** the **** ********** *********, **** ** * *****. ** **** breaks *** **** **** **** ******* **********, *** ***** *** similarities ****** *** (** ****) ** *** **** **** ** can *** ** ***** ** ************* ** *** ** ********* future ********* ** *** **** *******.

*** **** ******** ********* ****** **** ** **** ******** ** its *** **** ** **** ******* ** **** *** ******* learning ********* *** ******** ********** ****. ** **** *****, *** deep ******** ********* **** ** **** *******, ******** ****** **** humans *** *** **** ******* ** ******** ******, ** **** would **** **** **** **** ********* ** ******** ******, **** as **** *** ******** ************* ** ****** *************** ** ******.

What ***** ** "****"?

**** ******** ***** **** *** ****** **** ******* * ****** of ************ ************** ******, *********** *******, ** ***** *********. * system ******* ** ******** ******** ** ***** ***** **** ******* to ***** ** ****** ******** **** **********, **********, ******** ** badges ** *******, ****** ******, *** **** *****.

**** *** ****** ** ********, **** ****** ***** **** ******* them ******* ******** ****** (*********** ** *** ****** **** ** the ***** *****) ******* *** ******** ** *** ******* ***** various ******* *************, *** ****** * ******** ** ***** ***** of ******* *** **** ****** ** *** *****. ** **** case, "********" *** *** ******* *******, ** ** ******* *** most ********, **** *** ****** ****** ********** *** ***** ******** elements ********** **** "********" *** "*******" ******** ** * ****** extent:

***********, **** * ******** ****** *** ******** *** *** ****** to ** ********** ** "****", ******* ** ** *** ******** for ***** ** ** **+ ****** ** ************** ** **** advanced *******.

*** ** *** ******** **** **** ******** *** ** ***********, the ***** ** *** ******** ******* ** *** * **** indicator ** ******* ****** *********** ** ***********. **** **** *** imager **** ** * ****** ** *** ** ******** *********** factor ** ********** ** ***** *******.

How **** ******** ***** *********

**** ******** *** * ****** ************* ** *** *************** **** define ******* *******, ******* ** ******* ** ************ ** ***** appearance. **** ***** ** ****** **** ** ******** ******* ********** in *********** **********, ** **** *** ****** **** *** ***** any ***** *********** ** ************.

********* ************ ********* ******* ****** ********* ** ****** ** * time ** ***** ***** ** ****** *** ** (*.*.: *********** detection *****-*****). ** **** *****, ********* *** **** ** ****** "abnormal" ********, ****** ******* **** **** ******* ****, ** * large ***** *** ****** ** *** **** ********* ******** (**** as ** * ***** **********). ***** **** **** ***** ****** is ** *** **** ***** *****, ********** ******* **** * blob ** ****** ** * *****, *** *** * *** (or * *****, ** * ****). ** ***** **** ** better ********* ******* ** ******** ** * *****, **** ******** helps ***** ********* ******** ***** ******** *** **********.

The **** ** ********

* **** ****** *******'* *********** ** ***** ** *** ******* and ********* ** ****** **** *** ********. ******* * ****** that *** ******* ** ********* ****** **** ** ******* ** images ** **** ********, ** **********. **** ****** ***** ****** fail ** ********* * ******** ** ******** ******, ******* *** training **** *** *** ****** ******. ** ****, ** ***** perform ***** **** * ******** ********** ****** ******* ******** ******.

*******, **** *** ******** ** ***** ***** **** **** *********, or **** ** ********* ** *************** ** *** ******* ******** for **************. ***** ******** ****** ****** ******* ***** ** *** object **** ******** ******, *** *** ***** **** ******, ** a ******* ** ***** *** ********.

******* * ********** ** ******, *** ******* **** **** * category *** ** **** **** *********.********** *** ******** ********* **** *** *** ********, ** ** contains ******** ** ****** *********** **** ********* ** ********** *** sub-categories. ***"******" ************* ******** ****** *,*** ***-********** ** ******, **** *************** **** "warrior" ** "*********".

***-***** ***** ********* **** *** ********** ** **** **** *****-******* object *************** **** *******-******** *******. *** *******, * *** *** be *** ****** ** ********, ********** ** "**** ********* ******" and "********" (** **** ** ******** ***** *************** **** *****, shotgun, ***.).

**** **** ********, ** ***** ** ********** ** ****** *** system ** ******** ******** ************, ***** ** ***** ***************, ********* impractical *** *******-******** *******.

*********, ** ***** ***** * ****** ***** ****** ** *** and *****, ***** *** ***, ** ***** * *** ******* of *********** ****** *** ***. ********* ** **** **** *** becoming ******* ** ****** ************ ** ***** ********* ** ******* customer ******** ** ****** *** **** ** **** ************** ** browse ********.

Gender *** ***

*** ******* ***********, ********* ** ************ *****, *** ** ****** applications ** ** ****** ******** *********** ***** ** ***** ****** and ***. **** *** **** ******** ****, ****** ****** ** men *** *****, *** *** *****, **** ****** *** ******** to ****** ***** *********** **** ** ****** ** **** **** have **** ******* **. ******** ***** *******:

Filtering ****** **** **** ********

*** ********* **** ** ******* ****** ***** ************ ** ***** deep ******** ** * '******'. *** ******* ** **** ******** every ***** **** * ************ ****** ***** ** **** ******** intensive. ******* ** ***** ****, *** ****** ***** *** *********** video ********* ***** ** ****** **** *** ********* ******* ** analyze, **** ********** **** ****** ****** ******* ** **** ******** to ****** ** *** ****** ******* ** ********, *** *******, a ******, ******* ** * ***, * *********, * ****** or * *****. *** *******, ******* ********* **** ******** ******** ****, ** ***** **** ******* **** *********:

Training **** *** *********

**** * ********* ***********, ***** ** ** *** *** * user ** **** *** *** ****** *** *******, ************* ********* will *** ******* ******** **** *** ********. ***** ***** *** data **** ** ******** ** ********** ** *** ************ *** adapted *** ***-***** ** ***** *******, ********* ********* **** **** would **** ***** ************* ** ********* ** ***** **** ** re-use *** **** **** *** ***** *** ******** ********.

Potential ******** ****** *******

**** ** ******* ** ******** **** *** ** *********** **** systems *** ******* ***** ** ****-***** ******* **** *** ***** to *** ********** ******, *** ****** **** ***** *** ******* will ** ********. ******** *** **** **** ********* ** ***** vs. ****** ** ** * ******** ****** **. ** ********** area, *** ***** **** *** *** *** ****** ** *********** in ***** ****** ***** **** * ****** ** ******. ********, a ****** ***** ** ******* **** ********* ***** ** ******** or ******* ********. **** * ****** ******** ** ******* **** people ** ********* ********, ******, ***. ***** **** ******.

Hardware ************

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

  • ******** *** ****** ******* **** ******
  • ********* *** ****** ******* ** * ******* **** * ****** or ********

**** ** ***** ******** **** ** *** *** ** ****, this ** ************ ********* ** *** **** ** *******/*********, ** they **** *** **** ************* ***** **** ****, ******* **** new ******** ** ********* ******** *** *** ********* *** ******** to **** **** **** * ******** *******.

******** ** **** *************** *********, *** ********* **** *********** **** designed *** **** **** **** *** **** **** ********, *** expensive, **** **** ** **** ** * ****** ** ********. Depending ** *** ****** *** ********** ** ****** **** *** training, **** ******* *** **** *****, **** ** ***** ** complete, *** ******* **** ** ******** **** ** ********. *** training ***** ******* * ***** **** *** *** *** ** use ** ******** *******, **** ***** ** ********* **** ***** relative ** *** **** ** *** ***** ****. ****** ** a ****** ******** ** **** *** **** ****-*********** ******** *****.

********* *** *** ** * ****** ** ******** **** * lower-power *** ** **** **** **** ******** ** **** *******. Here, *****-**** *** ** **** ******* **** *** ***** ***********, and *** ****** ** ******* **** *** ** ******** *** classified ** *** *****, ** *** **** ** ***** *** system ** ******** ** ******. ********* *********/********, ************** ******* *********** **** ******** *** ****-***** ************.********** *** ****-***** ********, *** **** ******** ***** ********* ***** designed *** **-***** ********.

Deep ******** ** ************ ********

******* ** **** ******** ***** ************ ******************* ********* ** *** ******** ******** ******** ******** **** **** level ** **** ******** ******** ** ************. ************ ***** **** facial *********** ** *********** ****** *******, ** *********** ****** ****** to ****** ******-**** ****** ************.

Evaluating **** ******** ********

** *** *** ********* ** ****** * **** ******** *******, IPVM ***** ******** ********* *** ******* ** ****** ** * location *** *********** ** ******* ** *** ******** ********** ** possible. ***** ************ ***** ***** **** ****-**** *********** *********, **** do *** ****** *** ****** ************ *** ****-***** ***********. *******, tests ***** ** ********* **** * ****** ** *-* *****, giving *** ****** ********** **** ** *** * ******* ** objects, ******** *********** ** ** ******** **** ***/***** ********** *** across * ****** **** *** **** * ***** *****-**** ******* lot ****.

Future ********* *** **** ********

********** ********, *** ***** ** *** ** *** **** ******** development ** ********** ** *** ***** **** ** ** ******* and ******** *****. *** ************ ********* ***** *** * ******* improvement, ** **** *****, **** ******** *******, *** *** ***** far **** *****. ************ *** ***** ***** **** ** ****, core ********, *** **** **** **** **** *** ******** **** we ********** *** ******** ********* ***** ** **** ****** ***** when ******** ** ***** *********** ** *** **** *-** *****.

***** **** ** ********* **** *** ***** ********* **** *** be ********* **** * **** ******** ******* *** ******* **** deploying * *********-********* ********. *******, ***** **** * **** ********* need, ************ ** ******** ***** **** ****** *** ******* *********, would ** **** ** ******** ******* ** ****** * ******** too ***** ** *** ********* ***** ** **** ********.

Comments (26)

******** ** ****** ******, Deep ******** **** **** become * ****** *** bigger ****** ** ** try ** ********* ****** in *** *********. ******* will ** **** ** monitor ******** ***** *** also ***** **** ** event ******. ******* ********** can ******* ** **** used ** ****-******* ****.

***** ********. ****** *** providing **.

*********** ******** *************** ******* vs **** ********.

*** **** * ***** with **** ******* ** the *******. **** *** article:

******* ******** **** ***-********** instructions ** ***** * computer ** ********* ** image ** ** ******, deep ******** ******* ***** things *** *************.

**** ** ** ****** like *** ******** ** machine ********. **** *********:

******* ******** ** * field ** ******** ******* that ***** ********* *** ability ** ***** ******* being ********** **********.

********* ** ** *************, Deep ******** ** **** a ******* **** ** machine ********, *** ** many ********** **** ***** automatically ** ********** ******** data. *** ************** ************* * **** ** machine ******** **** **** requires **** ** ******** data, *** ***'* ********** deep ********. * **** used **** ********** *** detecting ******* ****** ** a *****, **** ******* thousands ** ****** ** license ******.

************ ******* ** ******* Learning******* **** ** ******* learning ************ ** ************* data **** ** ***** analytics, ***** ** *** context ** **** ******.

********** ******* ** **/******* Learning/Deep ********** **** ********-****** *******. They *** *** ********* example ** **** ***** machine ******** ******** ****** products ***** ******** **** "hand ******" ** ***** to **** *** ******** process **** *********. **** hand ****** ** *********** helping ****** ******* **** in *** *** ** consideration *** ******** ** get ****** ******* *******.

** ** ****** ***, one ** *** **** best *********** ***** *** machine ******** *** **** years *********** ******, ****** ** ***** required * ***** **** of ****-****** ** *** the *** ****.People ***** ** ** *** ***** ****-***** *********** **** **** ********* ******* ** *** ******* ***** ******** ***** ** ****** ******* *** *******; ***** ********* ** ********* ** ** *** ***** *****; * ********** ** ********* *** ******* “*-*-*-*.” From all those hand-coded classifiers they would develop algorithms to make sense of the image and “learn” to determine whether it was a stop sign. [Emphasis IPVM]

**** ** ******* ** my ******* ** *** report ** ****** *** machine ******** ****** **** coarse ********** ** ****** that ** *** *** to ********* **** ******* in *** ***** ****** be ******* ********.

*** **** **** *** nvidia *******

******* ********** *** **** ***** is *** ******** ** using ********** ** ***** data, ***** **** **, and **** **** * determination ** ********** ***** something ** *** *****. So ****** **** ****-****** software ******** **** * specific *** ** ************ to ********** * ********** task, *** ******* ** “trained” ***** ***** ******* of **** *** ********** that **** ** *** ability ** ***** *** to ******* *** ****.

* ********* *** ********* you ****** ******** ***********. I ** *** *** it ** * ********** of ******* ********. * see ** ** ** example ***** ******* ******** only ***** *** ** far, *** **** ** some ***** ******** **** coded ************ ** ******* the ********. ** **** were * ********** ** machine ********, **** ** would ** ********* ***** methods ** ********* ********, such ** *** **** Cascade ********* **** ** see ********, *** *******.

** ***** * ******** vision ******* *** *** to ****** * ************ model ** *** ********** to ** ********. *** example, ** ********* ******* plates ** **** ** define ************** **** * license ***** ***** ****. The **** ********** ******* machine ******** ** ********* computer ****** *** ********** neural ******** ** **** in ********* ******** ****** this ***** **** ** explicitly ******* ** * researcher *** **** ** may ** ******* ** machine ******** (*** *******, the **** ******** ********** may ** ***** ************* via ******* ********). ** artificial ****** ******** *** model ** ******* ********** via * ******** *** (and ** ******* ** even ***'* **** **** features *** ******* *** choosen ** ***** *** task).

** *** *** ********* to ****** * **** learning *******, **** ***** strongly ********* *** ******* be ****** ** * location *** *********** ** similar ** *** ******** deployment ** ********.

**** ** * **** good ******. ******* ** we ** *** ******* know ***** ******** *** used ** * ******* then **'* ********** ** predict ** ***** ********** the ******* **** **** pretty ****. *** ** ready **** *** ******* will **** ** ********** training ** **** **** exact ****.

**** ***** **** ** explicitly ******* ** * researcher *** **** ** may ** ******* ** machine ********

***, *** * ***** argue **** **** ***** can ** **** ******* such **** ** ********, we ******** ********** ** not **** ** ****** mathematically **** * ******* plate ***** ****. *****, I **** *** ******* of ******* ******* **********. Feed ** **** ***** of ******* ****** ** be **** ** ******** data, *** ** **** learn **** ** ***** effort **** ** ** detect ******* ****** ** a *****.

*** *** *****. * model *** ** **** or **** *******. *** both ******* *** ********** learning *** ******** **** other. * **** ****** to ********* **** ***** neural ******** ** *** to **** **** *** more ***** **** * researcher ** * ********. For *******, ** ******* classifier *** ***** ******** that *** ****** **** are (******) ******* ** a **********. ** ************* neural ******** **** ***** features *** ********** ****** learning *******. ** ****** there *** ***** * lot ** **** *** a ********** **** **** neural ******** :)

***, ******. *** ** is *** **** * am *********** **** ***** on *** *********** ******* deep ******** *** **** traditional *******. ***** * disagree ** *** *** article ***** ** ****** the *********** ************ ** **** ***** traditional ******* *** ******** all ****** ********* **** learning. **** ** **** not *******, ** **** industry ** *** *****.

* ***** ** ** important ** *** *** terminology *****. *** *******, a ******* ****** ** entitled ** *** *** term******* ********** ***** ********* ********* without ** ***** ********* by *** ******* ** mean ********* **** **** it **.

* ******* ****** ** entitled ** *** *** term******* ********** ***** ********* ********* without ** ***** ********* by *** ******* ** mean ********* **** **** it **.

**** ** *****, **** this *************, **** ******* ********** their ******* ** ***** "machine ********" *** *****'* mention *** **** **** learning

...**** ******** ******** ****** and ******* ******** **** put **** * ...

*** ****'** ** ****** use **** ******** ** their *******.

*****, ********* *******! * agree **** **** *****: Machine ******** ******* "********." The ******** ***'** ***** to **** ***** ** analytics ** **** * would **** "***** *****," because ***** ** ** learning. ***-******* ****** ***** are ****** ******* ** images ** *****. ****** there ** * ***-**-***-**** correcting ***** **** ** validating ******* ****, *** this ******** **** ** used ** ****** ** update *** ******** *****, I ***'* *** *** "machine ********" *** ** an ******** **********.

*******, ***** *** *** not ************ ******** *** use ** "********** ************."

******* ********, ***** ******, however ** ** "******" by ****-****** **** ***** constraints ** * ******** in *** *******, ** as **** ****** **** the ******* ***** *******.

** ***'*, *** **** the ****** * ****** of ******, *********** ******* it **** **** **** all **** ** ****** ("these *** ******** ** people"), ** **** ****** the ************* *** *********** components ** *** ***, without ****** ** ***-****** anything.

******* ********, ***** ******, however ** ** "******" by ****-****** **** ***** constraints ** * ******** in *** *******

*** ** ********, *** always ** *****. *** definition ** ******* ******** is ******* **** ****, you *** ********* *** limitations ** **** *********** machine ******** ********** **** the ***** ********** ** machine ******** *** ** doing ** *** ********* some ******* ******** ******* that ******* ** **** hand ****** ** ***********. It ***** ** **** accurate ** *** **** "traditional ******* ******** ********** versus **** ********", ****** than "******* ******** ****** deep ********".

**** ****** **** *** license ***** *******

* ***** *** **** the ************ ***** **** he ** ********* ** is *** *********** *** created ** *** ******** developers *** *** ******** an *** ******, *** a ******* ************ ***** created ** *** ***********/********* of *** ********** *** built **** *** ********* itself. ***** ****** *** be ******* ****** ** detect ******* ******, ** cars, ** ***** ********* only ** *** ******** data. *****, * **** my *********** **** *** OpenCV ******* **********, * got ** ** **** quite ****** **** *** and ** **** ***** constraints ** ************ ***** was ********.

***** *** ******** ******** ********* ******* ******** ** **** ********, where *** ******* ******** ****** ** ***** **** ****-***** ********** to **** ***, *** **** **** **** *********** ** ********** for *** ******** ******* ** ***** *** *********** ** *******.

**** ** *** *******:**** ******** **. ******* ******** – *** ********* *********** *** need ** ****!************ **** ****** ** ******* ** * ******* ******** ****** designed ** ****** ******* ** ******:

** ** ***** **** ** * ******* ******* ******** *******, we **** ****** ******** **** ** ** *** ****** *** whiskers ** ***, ** *** ****** *** **** & ** yes, **** ** **** *** *******. ** *****, ** **** define *** ****** ******** *** *** *** ****** ******** ***** features *** **** ********* ** *********** * ********** ******.

***, **** ******** ***** **** *** **** *****. **** ******** automatically ***** *** *** ******** ***** *** ********* *** **************, where ** ******* ******** ** *** ** ******** **** *** features.

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

** ****’* *** ********** ******* ******* ******** *** **** ******** then? **** ******** ******************** ********, *** ***** * ******** ******* ******** ***** ***** need ** ** **** *** ** ****** **** ** ******** prediction (** ******* ** **** ****), * **** ******** ***** is **** ** ***** **** ** *** ***.

*** *******:

*** ********* **** **** ******** *** **** *********** ***** ** machine ******** ** **** ***** *** ****** **** ** ******* a ********** *** ** ******** ** ***** **** **** ***** predictions, **** ******** *** ******** *** ********** ******** ******.

*** *******, ** * ****** ****** ** ******** ***** ***** in * ***** ** ***** *** **** ** ** ***** be *** *** ********** ********, **** ** ***** *** ********. It ***** ******* ** *** ** ****** ***** **** ** can **** ** ********** *** ********* ******** ** ***** ** make ** *********** ********** ***** *** ******* ** *** ******.

*** *** ******* ** **** ******, ***** ** ***** *********/******* vision, ******* ******** ******* **** ***** **** *** ******* **** amount ** ****-****** ** ********** **** *** ****** **** ** determine **** ***-******* ** *** ***** ** ****** ******* *** learn ****. **** **** ** **** ****-******, **** ** *** need ** ** **** **** ****** **** ** **** ******** aspect ******, ** ******** ********** **** (** * **** ********** example).

**** **** ******* ******, ***** *** ********** ** ****, *** we ***** ***** **** **-***** ******* ** ***** **********, ***., but ** ***** ** ****** *** ***** ** * '********', and ******** **** *********** ** *********** *** **********.

*****, *** *** ***** to **** * **** back **** *** ****. One **** **** **** this ********** ** **** it ******** *** ********'* loose *************, *** ***** vocabulary, ********* **** *** technologies. ****** * ****** more *********** ****** ** a **, ******* ** "Deep ********" *** ********* it ** *** **** greatest ***** ** ****** not **** **********. ** seems **** ***** ***** vendor ** *** & ASIS ** ******* **** learning, ** **, *** details *** **** ** exactly **** **** *** doing *** **** *********** how **** ** ****** works. *'* **********, *** cautiously **.

** ******** ***** ** that ** ** ********** ******** ****** **** learning, ** **** ******** is * **** ** machine ********, ****** ** should ****** ******** ************************ *********** **** *** *********** you ******** (********* **** coded **********) *** *** implicit ** *** ********** of ******* ********. *** titles ** ***** ******** are **** ****** ** the *********.

**** **********, ***, ** you *** ******** ** entire *** ****** ***** on *********** ******* ******** techniques, *** **** ********** have **** ****** ** the ******** *** **** special ***** **** *** have ** ***** **** for (*.* *******, *** bars *** ***** ***** obscuring *******, ********* ***** backplates ***) ******* *** deep ******** ******** ***** to ***** ********** ** well **** *** *** replace ***** ******* ****** with **** * ****** stage. *** **** ** work ****** ****.

"*******, ***** *** *** not ************ ******** *** use ** '********** ************.'"

*****... *** ****'* ******* what ************** ** ***** **** Learning ******* *****...

**** **** ******* *.*. once, ** *** ********* and ***** ********** **** learning *****. **** ********** it ** * ******* of ********** ************, ***** is *** *****.

"**** **** ******* *.*. once, ** *** *********..."

*** - '*****'.

** ******* *** ******* off ** ***** ****** in ***** ******** *** for******** *** ****. *** comedic ********* ***** **** everyone ********- ***** *** **** I *** **********.

********** ** ******* * guess. * ***** **** the ***** ******/******* ** the ***** *** "****":

"****'* ******* **** ************** ** ***** **** Learning ******* *****..."

*** **** *********.

** ** ******* * Deep ******** ***** ** the *** ***** ****** - ***** ** *********** followed ** *** ********** Intelligence ****** (***** ***** to ****** *** * long ****).

*.*. ***** '**** ********' youtube ***** '*****' **** the **** ******** ********* by ****.

**** ** *** ** opinion ** ** ********* with, *******, ** ** simply **** **.

**** ****** ***** *********** to ** **** ** not ********, **'* ** initial *** ** ********** that *** **** ****** against ***** *** ********** (to ****** * ********** degree ** *********). ** these ******* ********** *** defined ** * ******, and ****** ******* ***** without *** ******** ** modification, ***** ** ** learning. ******* ********, ** any **** ** ********, needs ** **** * feedback (****** ******** ** negative) ** ****** *** learning.

**** ** ***** * *** *** ** ********** *** ********** between *** *** ***'* **** ******** ** ***'*

*****://*****.**/-**********

** *** **** ***** of ************** **** ********** in *****, *** ** was ********** ** ***** that ** **** *** concept ***** *** ** reality *** **** ******* in *** **** **** future.

**** ** * **** useful ******** ** ******* explain **** ** *** general ******** ** ******* learning, **** ********, *** AI. ** ******** ******** our ******* ************ ****** with ****** ****** ******* AI ** **'* **** intriguing ** ***** **** about *** ********* ** what **** *** *******.

Login to read this IPVM report.
Why do I need to log in?
IPVM conducts unique testing and research funded by member's payments enabling us to offer the most independent, accurate and in-depth information.

Related Reports

Verint Victimized By Ransomware on Apr 18, 2019
Verint, which is best known in the physical security industry for video surveillance but has built a sizeable cybersecurity business as well, was...
Door Operators Access Control Tutorial on Apr 17, 2019
Doors equipped with door operators, specialty devices that automate opening and closing, tend to be quite complex. The mechanisms needed to...
Axis Supports HD Analog on Apr 15, 2019
In 2017, Axis declared 'Everything is IP': Now, in 2019, Axis has released support for HD analog, with their new encoders.  Why the change?...
ISC West 2019 Report on Apr 12, 2019
The IPVM team has finished at the Sands looking at what companies are offering and how they are changing their positioning. See below for 50+...
Pole Mount Camera Installation Guide on Apr 11, 2019
Poles are a popular but challenging choice for deploying surveillance cameras outdoors. Poles are indispensable for putting cameras at the right...
Bosch AI Camera Trainer Released And Tested on Apr 09, 2019
Bosch is releasing a highly unusual new AI feature - 'Camera Trainer'. Now, coming as a standard feature in Bosch IVA/EVA analytics, one can train...
Airship VMS Profile on Apr 03, 2019
Airship has been developing VMS software for over 10 years, however, with no outside investment, and minimal marketing, the company is not well...
IBM / Genetec Surveillance System Investigated Over Philippines Human Rights Abuses on Mar 22, 2019
A lengthy investigation into an IBM video surveillance project in the Philippines, raising concerns IBM helped local police conduct a bloody...
Large Hospital Security End User Interview on Mar 21, 2019
This large single-state healthcare system consists of many hospitals, and hundreds of health parks, private practices, urgent care facilities, and...
Retired Mercury President Returns As Open Options President on Mar 18, 2019
Open Options experienced major changes in 2018, including being acquired by ACRE and losing its President and General Manager, John Berman who...

Most Recent Industry Reports

H.265 Usage Statistics on Apr 19, 2019
H.265 has been available in IP cameras for more than 5 years and, in the past few years, the number of manufacturers supporting this codec has...
ACRE Acquires RS2, Explains Acquisition Strategy on Apr 19, 2019
ACRE continues to buy, now acquiring RS2, just 5 months after buying Open Options. One is a small access control manufacturer from Texas, the...
Access Control Course Spring 2019 - Last Chance on Apr 19, 2019
Register for the Spring Access Control Course. IPVM offers the most comprehensive access control course in the industry. Unlike manufacturer...
Riser vs Plenum Cabling Explained on Apr 18, 2019
You could be spending twice as much for cable as you need. The difference between 'plenum' rated cable and 'riser' rated cable is subtle, but the...
Verint Victimized By Ransomware on Apr 18, 2019
Verint, which is best known in the physical security industry for video surveillance but has built a sizeable cybersecurity business as well, was...
Milestone Drops IFSEC on Apr 18, 2019
Milestone has dropped out of Europe's largest annual security trade show (IFSEC 2019), telling IPVM that they "have found that IFSEC in EMEA no...
The Fastest Growing Video Surveillance Sales Organization Ever - Verkada on Apr 17, 2019
Verkada has the fastest growing video surveillance sales organization ever. In less than 2 years, they already have more salespeople in the US...
Door Operators Access Control Tutorial on Apr 17, 2019
Doors equipped with door operators, specialty devices that automate opening and closing, tend to be quite complex. The mechanisms needed to...
Securadyne CEO: IPVM 'Entertaining For An Ignorant Few' on Apr 16, 2019
Securadyne's CEO Carey Boethel is unhappy with IPVM's report - Failed Integrator Rollup, Securadyne Sells to Guard Giant Allied. Indeed, he...
Dahua Repositionable IR Multi-Imager Camera Tested on Apr 16, 2019
Dahua has released their first repositionable multi-imager camera, the Multi-Flex 4x2MP, claiming integrated IR, true WDR, and flexible...

The world's leading video surveillance information source, IPVM provides the best reporting, testing and training for 10,000+ members globally. Dedicated to independent and objective information, we uniquely refuse any and all advertisements, sponsorship and consulting from manufacturers.

About | FAQ | Contact