dschmiconti
New member
Hi all,
as the title of this thread suggests I am interested in the performance impact of running multiple neural networks in parallel on one NPU.
Let's take this as a scenario:
* Face detection on an near-infra red camera
* object detection and classification on two separate RGB-cameras (they are facing in different directions, but the NN i want to use is the same for both camera images)
* keyword detection and subsequent natural language processing on a microphone input
In the system the applications would be running in parallel, and not necessarily know about each other.
How would that impact the performance of the system? Is there a benchmark for such a scenario?
Any input from your side would be appreciated. Thank you in advance.
Best regards
Daniel
as the title of this thread suggests I am interested in the performance impact of running multiple neural networks in parallel on one NPU.
Let's take this as a scenario:
* Face detection on an near-infra red camera
* object detection and classification on two separate RGB-cameras (they are facing in different directions, but the NN i want to use is the same for both camera images)
* keyword detection and subsequent natural language processing on a microphone input
In the system the applications would be running in parallel, and not necessarily know about each other.
How would that impact the performance of the system? Is there a benchmark for such a scenario?
Any input from your side would be appreciated. Thank you in advance.
Best regards
Daniel
