Multi-stream processing using eepStream on Jetson Nano

Hi all,
I see deepstrean on jetson nano, it’s very good job.
I have some questions about demo.

Q1- I want to know, which of detection model(s) are used for this demo? It’s likely to be the ssd model, but I want to know, Is it the ssd mobilenet v1/v2/v3 or custom ssd with extremly small backbone like resnet 10/18?

Q2- How many the model is used in this demo package? only have one detection model for all 8 streams? and this detection is detect only a few object class or coco dataset class?

If you are using DS 5.0, all the models under /opt/nvidia/deepstream/deepstream-5.0/samples/models.

DS 5.0 is open source and free?

No, but all the source code of deepstream-app is public.

Is it possible to run these source codes on jetson nano and custom models in python language?

Yes, I think you need to refer https://docs.nvidia.com/metropolis/deepstream/dev-guide/index.html firstly and many problems will be solved.