- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I can run the C++ executable of "object detection sample ssd" with Intel's person detection model (person-detection-retail-0013) with no problems. If I run it with the Python script of object_detection_demo_ssd_async.py with just input and model arguments I get:
[ INFO ] Reading IR...
[ ERROR ] Following layers are not supported by the plugin for specified device CPU:
detection_out, mbox1/priorbox
[ ERROR ] Please try to specify cpu extensions library path in demo's command line parameters using -l or --cpu_extension command line argument
If I add the cpu extension library argument for cpu_extensions.dll, the plugin load line (exec_net = plugin.load(network=net, num_requests=2) causes the process to terminate. No exception, the Python process just terminates.
Obviously I am missing something but at this point I am out of ideas. Anyone have any insight to what is going on?
Link Copied
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
It dawned on me that I had seen this behavior before with Python and OpenVino. It occurs when the the inference engine dll is a release version ant the dll you are attempting to use is a debug version. I rebuilt the cpu extension dll as a release version and problem solve.

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page