How to use different models on different sources using Python pipelines?

Accelerated Computing Intelligent Video Analytics DeepStream SDK
#1

Please provide complete information as applicable to your setup.

• dGPU
• 5.1

Is there any sample code where:
1.For Stream/Source 1 & 2 I use detector A
2. For Source 3 & 4 I use detector B?

Basically this will involve running two concurrent pipelines I would imagine? Would this be possible using a single python script?

#3

Below is all the samples.

https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_Python_Sample_Apps.html#id1

#4

Hi, thanks for the reply, but there’s no example showing this.

What I want to do is run different pgie and sgies on different input sources. E.g. a vehicle detector on source 1 and 2 and say a bed detector on source 3 and 4.

Note, the detectors can be combined, but they’re in vastly different domains so ideally I’d like to keep them separate. Is that something DeepStream supporters?

It’s basically two different pipelines in the same app.

#5

Yes. DS can support two different pipelines in the same app.

#6

Hi, Kesong,

Great, do you have any example code showing this?

#8