Subscriber Discussion

Video Analytics Accuracy Differences - RTSP Vs Recorded Stream?

Hi,

I am running wrong movement detection model, when i am giving the recorded video (which is in mp4 format) as input we are getting very good accuracy, where as when we are checking direct camera RTSP we are not getting the results. there is no issue in the RTSP streaming, since we can see the proper recorded video and also for other analytics ex: no parking, we are getting very good accuracy.

* ********** **** *** ***** ** mainly ******, ***** **** ******** ** the ****** ** ********.

** *** ******** *** ******** ***** on *** ****** ***** ******** ******** vs *** ****** ******** ***** ******** vehicles, ****** ***** ******** ******** *** counted ********, ***** ** ****** ******** wrong ******** ******** ** ***** ** the ****** ********* ******

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

Agree
Disagree
Informative
Unhelpful
Funny
Login to read this IPVM discussion.
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.

***********. ***** ********* *** *** *****? And *** ************** *** ******* *** same *** **** ***** *******? *** recorded *****, *** ** **** ***** input? ********* ** *** **?

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

**** ****, *** ******** ** ****** to ******* *** ***** ** **** time *** *** ** ******** ******, leading ** ***** ******* ****** ******** video.

***** *** ** **** ************* ** setting *********** *** *** ********* *****.

Agree: 2
Disagree
Informative
Unhelpful
Funny

**** ** *** *** ********* ***** development *****.

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

***** ** * ****** ** **** video *****, **** **** * ** giving *** ***** ***** ** *** model.

** *** ***** **** ** *** streaming *** ** *** ***** ****** recorder *** ***** ********* ******. *** same *** *********** *********** *** ****.

* ** *** ****** *** ***** loss, ** *** ***** *** **** we *** ****** *** **** ***** also, ** *** *** ****** *** jittering. ***, ***** ********** *** **** to ***** ***** *** ** **** drops, *** ** ***** *** *** to ***** ****.

** *** ** ************** *** ******* there ** *** ********** ******* *** modes.

Agree
Disagree
Informative
Unhelpful
Funny

*** ******* ******** ***** **** ********, but **** **** ******** * ** guessing **** ** **** ***** *** lines ** **** ******** ****-****** ****** and *** ******* ***********.

**** *** ***** ** ****, *** streaming **** *** ***** **** *** (or ******** *** ******) ** **** streams. **** **** ***** *** ******** to ******* **** ******* *** **** pipeline ** ** **** ****** ******.

** ***** ** **** ******** ******** you *** ***** ********* ******** ***** differently **** ****/**** *****. ******* ****** results ** **** ***** **. ****** input ** *** ********, *** * variety ** *******.

Agree: 2
Disagree
Informative: 1
Unhelpful
Funny

**

*** **** *********** **** *** ****** RTSP ** ***** ** ***** ** model, ** *** ******* *** ** jittering, **** ** *** *** ****** video.

***** ** **** *** **** ****** is ********* ** *** ** ***** and **** *** **** ** ***** from *** *** ***** ** ***** to ***** *** ******* *** *********. As *** ** ********* ** *** will ** *** **** **********. **** will ****** **** *** **** *** then **** ** ** ** ***** analytics ******, ***** ********* ******, **** exactly ******* ** **** ********* ****** level.

**** ** * *********** *** ********* concept ****** *** ***** *.*. **** then ********* **** ** *** *****, can ** ** ******** *** ****?

Agree
Disagree
Informative
Unhelpful
Funny

**

*** **** *********** **** *** ****** RTSP ** ***** ** ***** ** model, ** *** ******* *** ** jittering, **** ** *** *** ****** video.

***** ** **** *** **** ****** is ********* ** *** ** ***** and **** *** **** ** ***** from *** *** ***** ** ***** to ***** *** ******* *** *********. As *** ** ********* ** *** will ** *** **** **********. **** will ****** **** *** **** *** then **** ** ** ** ***** analytics ******, ***** ********* ******, **** exactly ******* ** **** ********* ****** level.

**** ** * *********** *** ********* concept ****** *** ***** *.*. **** then ********* **** ** *** *****, can ** ** ******** *** ****?

Agree
Disagree
Informative
Unhelpful
Funny

***** *** *** ********* ***** *** differences ******* *** *** ****** *****, I ******* ******* *** ****** **** set ***** ******** *** **** *******.

** ** ********** * ****** **** set ** ***, *** ** ****** good *******.

**** ** ** ******* ******* ** parse * *** ****:

******* -*********** -* "*:\****\****-****.***" > ****-**-*****.***

*** ***** **** **** ** *** txt **** *** *** *** *********:********=*** *** **** **.

***** ** **** * ***-**** **** shows *,* ** * **** *** be *********** **** ****.

******* ******* ** *** **** * frames *** ***** ***** **.

Agree: 1
Disagree
Informative: 4
Unhelpful
Funny

***** * ****** ******** *** ********* video ** * ******** **** ** you *** ***** ** ** ***** that ***** ** ***** ** ***** analytics. *** *** *** *** ** record **** *******.

Agree
Disagree
Informative
Unhelpful
Funny

** *** *** *** **** *****, but **** ** *** ***** ***, when **** ** ***** ******* *** model, ** *** *** ***** ** the *** ** *********. **, ** started ***** **** ** ** ****** the *****

Agree
Disagree
Informative
Unhelpful
Funny