2

I want to load and run multiple networks on the same NCS2: a one-class object detection network (like a person detector), and a network for some recognition on that detection (like gesture recognition). I tried to load the networks on one NCS2 through two different threads. But when loading the second network, the program exits without any warning or error; the networks are separately working fine (one at a time).

I am using Python on Raspberry pi 4/Raspbian Buster, and the networks are in IR (xml + bin) format.

  1. Is it possible to load multiple networks on the same NCS2 at all?
  2. If yes, what do I miss? Do I have to do some configuration or so?
Fardo
  • 107
  • 1
  • 1
  • 9

1 Answers1

4

Yes. It is possible. No specific configuration actions required.

There are examples of such functionality in a repo open-model-zoo. For example, this one. Action recognition demo based on two networks. The demo is implemented using Python.

Any chance to share source code of your app? It would be a lot easier to understand what may go wrong.

Artemy Skrebkov
  • 346
  • 1
  • 8
  • Dear Artemy, Thank you very much for your quick and accurate response. Actually, based on your answer, I was able to find the problem: I was trying to load two models on the same NCS2, but one of them was implemented using Opencv (i.e. "cv2.dnn.readnet" method), while the other one was implemented directly using Openvino Python API. I tried and implement the first one using Openvino API and the error is gone. – Fardo May 23 '20 at 21:53
  • Great! Yeah, when you mix APIs which use the same backend weird things might happen :) Glad that you could resolve your issue. – Artemy Skrebkov May 24 '20 at 15:34