Cannot run inference on OpenVino CPU after update to OpenVino 2022.2 #22640
Description
System Information
OpenCV version: 4.6.0
Operating System / Platform: Win11 x64
Compiler & compiler version: MSVC 2022
Detailed description
We've been inferring a resnet based dnn for years using OpenCV, with support for different backends.
Recently, after upgrading to OpenVino 2022.2 and recompiling OpenCV 4.6 we have an exception when inferring on several Intel CPUs (one i9-gen10, one i5-gen8) (it works on Intel GPUs):
Exception: OpenCV(4.6.0-dev) D:\Dev\opencv\modules\dnn\src\ie_ngraph.cpp:747: error: (-2:Unspecified error) Failed to initialize Inference Engine backend (device = CPU): Cannot get memory! in function 'cv::dnn::InfEngineNgraphNet::initPlugin'
This might be a bug for https://github.com/openvinotoolkit/openvino/ as the exception is trigger here:
opencv/modules/dnn/src/ie_ngraph.cpp
Line 1070 in 3472469
but with a message from
https://github.com/openvinotoolkit/openvino/blob/042bd7274ac36715e16386be5c1f17924d53dede/src/plugins/intel_cpu/src/cpu_memory.h#L206
Steps to reproduce
try to call forward() on a DNN model
Issue submission checklist
- I report the issue, it's not a question
- I checked the problem with documentation, FAQ, open issues, forum.opencv.org, Stack Overflow, etc and have not found any solution
- I updated to the latest OpenCV version and the issue is still there
- There is reproducer code and related data files (videos, images, onnx, etc)