Segmentation C++ 2020R4 crashes

[Tigi_T_Intel]

Segmentation C++ 2020R4 crashes

Segmentation C++ 2020R4 crashes during the inference. 

segmentation_demo.exe -m "C:\\Temp\\Tigi\\OV2020Models\\Unet\\FP16\\unet-camvid-onnx-0001.xml" -i "C:\\Temp\\Tigi\\Semantic Segmentation\\Images\\3.jpg" -d "CPU" 

The execution stops after hitting InferRequest.Infer(); in the code. 

I have tried GPU and that also fails. 

When i have tried python version, it works fine.. 

[Tigi_T_Intel]

Segmentation C++ 2020R4 crashes during the inference. 

segmentation_demo.exe -m "C:\\Temp\\Tigi\\OV2020Models\\Unet\\FP16\\unet-camvid-onnx-0001.xml" -i "C:\\Temp\\Tigi\\Semantic Segmentation\\Images\\3.jpg" -d "CPU" 

The execution stops after hitting InferRequest.Infer(); in the code. 

I have tried GPU and that also fails. 

When i have tried python version, it works fine.. 

[Iffa_Intel]

Greetings,

1.If possible, could you provide the link to your project resources, such as your models so that I could replicate the same?

2.Are you using Windows or Linux?

3.Have you setup your Openvino toolkit as in here: https://docs.openvinotoolkit.org/latest/openvino_docs_install_guides_installing_openvino_windows.html

4.Besides, any chance that you are referring to our official documentation which is this one: https://docs.openvinotoolkit.org/2018_R5/_samples_segmentation_demo_README.html

5.So far Windows CPU has no problems to run the car_1.bmp from this resource https://download.01.org/opencv/2020/openvinotoolkit/2020.4/open_model_zoo/models_bin/3/unet-camvid-onnx-0001/FP16/

Sincerely,

Iffa

[Tigi_T_Intel]

Hello,

Segmentation works well for Release mode.

But it has issues on the Debug Mode..

Regards,

Tigi

[Iffa_Intel]

Greetings,

Fyi,

the pipeline of object_detection_demo_ssd_async.py will be redesigned soon

You can refer to this link for PR request:

https://github.com/openvinotoolkit/open_model_zoo/pull/1236/files/8442eb70555b75a2344da4cbb60361187ace4ff3

Sincerely,

Iffa

[Iffa_Intel]

Greetings,

Intel will no longer monitor this thread since we have provided a solution. If you need any additional information from Intel, please submit a new question. 

Sincerely,

Iffa