Intel® Distribution of OpenVINO™ Toolkit
Community assistance about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all aspects of computer vision-related on Intel® platforms.
6430 Discussions

rtsp very slow in face_recognition_demo from openvino 2020.3

aicha
Novice
1,113 Views

hello intel,

I am a student just starting to use openvino toolkit and trying to run the face recognition demo from the open model zoo . the demo runs fine when using the pc camera .. but it is wery very slow when I use rtsp of an ip camera .. please help me solve this. 

Thank you in advance .

0 Kudos
1 Solution
Syamimi_Intel
Moderator
965 Views

Hi Aicha Ezzahra Sassi,

Please check your data rate excluding inference workload. This issue might be due to a slow incoming data rate.

Other than that, this issue might be due to using NCS2 as an inference device which can slow the output rate. You can try run the face_recognition_demo using a different device.

 

 

Regards,

Syamimi


View solution in original post

0 Kudos
7 Replies
Vladimir_Dudnik
Employee
1,075 Views

data rate of rtsp or IP connection might be slow, obviously demo can't work faster then incoming data rate

0 Kudos
aicha
Novice
1,054 Views

hello Vladimir, thank you for your response.

The demo works very well when I use usb camera. Do you have any solution for the slow performance of the rtsp ? or any alternative to it ?

( I use i7 8th gen intel processor and intel neural compute stick 2)

I will be greatefull to you if you can help me with that.

0 Kudos
Vladimir_Dudnik
Employee
1,041 Views

data rate of rtsp or IP connection might be slow, obviously demo can't work faster then incoming data rate

0 Kudos
Syamimi_Intel
Moderator
1,015 Views

Hi Aicha Ezzahra Sassi,

I have validated face_recognition_demo using an IP camera and it works fine. For your information, I am using the IP Webcam application to run the demo. 

The demo run slow maybe because data rate of RTSP or IP connection is slow.

 

Regards,

Syamimi

 

0 Kudos
aicha
Novice
1,002 Views

hello @Syamimi_Intel ,

thank you for your response. But I didn't find any solution for that till now .

regards.

aicha

0 Kudos
Syamimi_Intel
Moderator
966 Views

Hi Aicha Ezzahra Sassi,

Please check your data rate excluding inference workload. This issue might be due to a slow incoming data rate.

Other than that, this issue might be due to using NCS2 as an inference device which can slow the output rate. You can try run the face_recognition_demo using a different device.

 

 

Regards,

Syamimi


0 Kudos
Syamimi_Intel
Moderator
916 Views

Hi Aicha Ezzahra Sassi,

This thread will no longer be monitored since we have provided a solution. If you need any additional information from Intel, please submit a new question.



Regards,

Syamimi


0 Kudos
Reply