- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I know this has been asked before, but after trying to implement suggested solution it did not work out..
I have tried running the human pose estimation demo on the raspberry pi 3B+ model and got a throughput of 2.77 FPS. So I tried using 2 NCS 2 devices on a normal machine running Ubuntu (virtualBox) and got the performance of 3.44 FPS max.
So my question is:
Does using multiple NCS 2 devices actually help the performance of running the model and how to enable it do so?
Could you explain how we can modify the human pose estimation demo to be able to run on multiple NCS 2 devices.
Thank you,
Sweta
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi SDwiv1,
The developer forum (https://ncsforum.movidius.com) will be migrated to this forum. It is possible to use multiple NCS device with the OpenVINO Toolkit. The Inference Engine API will distribute inferences to the NCS devices based on the device load. Take a look at "Multiple NCS Devices" section at the following page for additional information.
https://software.intel.com/en-us/articles/transitioning-from-intel-movidius-neural-compute-sdk-to-openvino-toolkit
Regards,
Jesus
- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page