Frames result in Multi Camera Multi Person Python* Demo

[Lenard-Lim]

May i know that when follow this demo step, two cameras will show the result in one frame or two frames?

[Sahira_Intel · Accepted solution]

Hi Lenard,

When you use multiple streams of video in this demo, you can specify the batch size which will take all the frames and perform inference on each frame at the same time.

For example, using the argument -bs x (where x is the number of video streams), the demo will grab a frame from each video stream and perform inference on each frame all at once.

I hope this information is helpful, please let me know if you have any further questions.

Best Regards,

Sahira

[Lenard-Lim]

Hi sahira,

So which mean that this demo will grab all possible video stream and perform to each difference frames. right?

I have another question for this demo: because of multi cameras, may i know that each camera need their movidius neural compute stick or multi cameras can share by one movidius neural conpute stick?

Thank you.

regards,

Lenard Lim

[Sahira_Intel]

Hi Lenard, 

Yes, you are correct.

And you can use 1 NCS2 to perform inference on multiple video streams. You can use multiple NCS2 if you would like to speed up inferencing, but it will work with just one. 

Best Regards,

Sahira 

[Lenard-Lim]

Hi Sahira,

Sorry for late reply.

Thank you for you reply and which is helpful for me.

May I know that is the Multi Camera Multi Person Python* demo have been not available. This is because I found back the link and the link is show 404.

Can you provide other link or any update link for multi camera multi person python demo?

Thank you.

Regards,

Lenard

[Sahira_Intel]

Hi Lenard,

Looks like the name was changed slightly, but here is the demo:

https://github.com/openvinotoolkit/open_model_zoo/tree/master/demos/python_demos/multi_camera_multi_target_tracking

Best Regards,

Sahira

[Lenard-Lim]

Hi Sahira,

sorry for the late reply and thank you your solution.

regards,

Lenard