Multi-Camera Tracking Works?

By Carlton Purvis, Published Jun 27, 2013, 12:00am EDT (Info+)

Being able to track people across cameras and buildings simply by using cameras and analytics is something of a 'holy grail' for the surveillance industry.

Two researchers from Carnegie Mellon University have developed just such a system that allows them use surveillance cameras to track a person's movements throughout a building, even when they are out of sight of the camera. The set-up combines three different datasets that allow them to tag and identify people and place their movements on a map. In this note, we interview one of the researchers and breakdown how the system works.

**** *****************-* *********** **** ******** ******* **** ******** used ** **** *****-****** ******** ******* include ** ***** *** ****** **** a ****** **** ** *** ***** environment, ***** ** *** *** **** in **** ************ ***********. **** **** especially ********** ** ************ ******** ****** in “******* ****** ************.”

*** ***** **** **** **** ***** in * ******* **** ** ****, with ****** ********* (*** *** ****** layout *****), **** ***** *** * moderate ****** ** **** ******* ****** the ***. ********* *** ** ***** that **** ***** ************ ***** ******, even ******* ****** **** ** ********* data **** ********, ********** *** ****** recognition. *********** *** **** ********* **** ****.

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

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

  • *********** ********** ******* **** ***** ********. *** FoV ***** *** ******* ****** **** 30 ** ** *******.
  • *** ****** *** ***** **** ******* -- “We ******* [*******] **** ***** ** 15 *** ******, *** **** ** to * ******. ******** *** ***** just **** *** ******** ** *** cameras *** ** ****'* **** ****," Hauptmann ****.
  • ******** ****** *********** ******** (****** ** was****** ** ******)
  • * ******** ********* ******* ** ******* clothing *****, **********, * ***-****** ** map ** *** ********, *** ****** recognition.

********* ****, “** ****’* ******** **** this [*********] ** ******* ****** *********** or ****** ********* ** **** ** recorded, ** ******** ** ** **** a ********** ** ******** (*******), *** this ** **** ** ***** ** with. *** **** ***********, ** ****** reasonably ****.” **** **** **** ** track * ******* ****** *** ***** 88 ******* ** *** ****.  

Multiple **** *******

******** *****

“******** *** ***** ***** *********** **** have ********** ******** ******** ******** **** multiple ****** **** **** ***** ******** and ***** **. ********** ***** *********** will **** ** ************** ********* ******,” according ** ***** ***** ******. *********, color *** ***** ******* **** ********* to ***** * ****** ********** *** facility.

******** ******* ****** ****** ** ****** lighter ** ****** ** ********* ******* so *** ****** *** *** ** allows ******* ** **** ***** *** color *************. ** ** ****** ****** a ******* *****, *** ****** ***** still ********* **. ******* ******** ******* **** the ******* **** ********* *** **** “they ***’* ***** ** *********** ********* colors,” ********* ****. ***** **** ** the ********* **** ******* ******* ******, giving *** ****** * ********* ** distinguishing ******* ****.  

** **** ***** *** ********* ***** use ********** ** “******” ** *** the **** ****** ** ******. “******* you **** ** ** ***** *** think *** ****** **, *** *** camera **** ‘** *’** *** *** person ** **** *** *’* ****** this ***** ************* *** **** ****** now.’" If *** ****** ******** ** ****** wearing ******* ******, *** ********** ********* would **** ** *** ** ******** higher *** **************. **** ***** ***** * mutual ********* ********** -- * ****** cannot ** ** *** ****** ** one ****.

**********

********** ********** ***** ** *********** ** where * ****** ****** **, ***** on ***** ********* ** ******, * 3D ************ ** *** ******** **** the ********* “****” *** ***** *******. Even **** * ****** ** *** contributing *********** ** * ******, ** its *******, *** *** ***** *** if * ****** ** ** *** opposite **** ** * ****, ********* by * ****** *******. ************ ******* **** three ******* **** **** *******. *** limitation ***** ** *** ***** ***** be ********* **** ******* ** *** camera. 

***** ****** **** *** ******** **** person ** *** ****** *****. *** camera ******** **** ***, *** ***** is ********** ***** ** ***** ****** views *** *** **********.

****** ***********

****** ********* *** ********* **** ** supplement ******** **** * ****** ***** not ** ********** ** ******** ***** or ********* **********. ** *** **** the **** **** ********* (****** ***** six ******* ** ********* * ****). Additionally **** ****** ****** **** **** suitable *** ***** **** *****, *** because “**** ** *** ******** **** older ****** ** ** **** **** walking **** **** ***** ******* ****.” When ******* *** *** * ***** shot ** * ******’* ****, **** data ***** ****** **** ** *** the **** ****** ******** ** ********** and ******** *****.

“*** **** ** [****** ***** ****** recognition] *** *** **** **** *** there *** ****** *********** ****** ** to *** ** **** ** *** this ****** *****,” ********* ****. **** 10 ******* ** *** ** *** video ****** ********* ******** ****** ****** information, *** “** ******** * ********* of ***** ********** **** *** ****’* get *********. ******* ****** *********** *** ****** could ***** * ****** ****** *** ***** 58 ******* ** *** ****. *** algorithm ***** ****** *********** *** **** to ***** * ****** ****** *** meter ** ******* ** *** ****. 

** *** **** ** ** ***** direction *** **********, ** * ****** stops ** **** ********, *** **** almost ** ****** ** **** ***** of ****. *** ****** *********** ***** us **** ** * ********** **** says **** *** ** ** **** person **** ***** ***.”

*** *********** **** ******* **** *** algorithm ** “*** ********* ** **** crowded ********* ***** **** ****** ***** the **** ***** *******,” ********** ******* of ****** *********** **********, ** ******** trajectory (****** ****** *** ****/*** *********** to ******** * ****) *** *** much ** *** **** *****.

**** ******* ***** ******* **** **** at *********** ****** *** ******* *********** **********.

Future *****

"** **** ******* ******** *** ********* substantially ******* – **** ******* ** ongoing ** ***** ** ****** *** technology. Currently ***** *** ** ******** ***** to ****** *** *****. **** ***** need * *********** ****** ** *********** and *******," ********* ****. 

***** *** ** ***** ** ************* the ******** (********* ** ********** ****** it ** ***** ** *** ******** and *** **********), *** ** ***** ** **** available *** *********, ********* ****. 

IPVM ********

********* **** ******* ***** **** *** application **** *** ** **** **** they ******* ***** ********. *** ******** system ***** ** * *** ** keep **** ** ******** *** ***** wander *** ** ** **** ***** people *** ** **** ** * medical *********. **** ********** ***** ** beneficial ** ************ **** ******** ***** where ****** **** ** ** ********* for ****** *** ****** *******, *** also ** **** ******** *****. *** first *********** ** **** **** **** to **** *** ****** ********** **********. The **** ****** *** ******* *** spent ******** ** ****** ** ****** control ** **** *** ** ***** to **** ***** ** *** ********. Hauptmann's ********** ***** **** **** ********** ability ** **** ******* ***** **** are ** * ********, *** **** what ****'* ****'** **** *******. ******* ********** is ************** ** ******* ********** **** ***** ** ****** ** ******* and **** ***** ***** ************ ** track * ****** ** ********. 

** *** ******** ****, *** ***** density ** ****** ******** - ** across * **** ***** *****, ******** significant **** *** ********** ***********.

*******, ******** ****** *** ***** ********** from **** *** ***** ********* ** 2005, ** ** ***** ** *********** to *** *** **** ***** ***** turn *** ** ********** ** **** with ******* ********** *** ****** ********** power.

Comments (3)

Isn't that what camera mapping is for?

Very complexed and commercially not ready perhaps ...

These guys - 20220surveillance.com - (whom I've never heard of) claim to have an operational model of multi-camera tracking.

Screenshot from the video on the splash page above:

Login to read this IPVM report.
Why do I need to log in?
IPVM conducts reporting, tutorials and software funded by subscriber's payments enabling us to offer the most independent, accurate and in-depth information.
Loading Related Reports