frame

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

Sign In

Howdy, Stranger!

It looks like you're new here. If you want to get involved, click one of these buttons!

[ ERROR ] AssertionFailed: inputs.size() == 1

Hi,

I am setting up raspberry PI along with the Movidius NCS stick. I am referring to the setup in https://docs.openvinotoolkit.org/latest/docs_install_guides_installing...

When I try to run the example face detection model, I get the following error.

[ ERROR ] AssertionFailed: inputs.size() == 1

When I try the OpenCV example, the error looks like this

cv2.error: OpenCV(4.0.1-openvino) /home/jenkins/workspace/OpenCV/OpenVINO/build/opencv/modules/dnn/src/op_inf_engine.cpp:553: error: (-215:Assertion failed) Failed to initialize Inference Engine backend: AssertionFailed: inputs.size() == 1 in function 'initPlugin'

Can someone help me?

Praveen

Comments

  • 1 Comment sorted by Votes Date Added
  • Hi @tpp6me,

    Thanks for reaching out. It looks like you may have missed initializing the OpenVINO™ Toolkit environment variables, I was able to reproduce the problem and initializing the variables fixed it for me. Execute source /opt/intel/openvino/bin/setupvars.sh. In the Install OpenVINO™ toolkit for Raspbian* OS guide under the Set the Environment Variables section, there is an Optional step to permanently set the environment variables so that you don't have to worry every time a new terminal window/tab is used.

    Also the openvino_fd_myriad.py sample code found under the Run Inference of Face Detection Model Using OpenCV* API section, in the sample code, line 10 frame = cv.imread('/path/to/image') make sure you add the path to your image (i.e. '/home/pi/Pictures/face.png'). Don't forget to also include the network files (i.e. 'face-detection-adas-0001.xml' and 'face-detection-adas-0001.bin') in the same directory as the openvino_fd_myriad.py file. Let me know if the issue persists, hope this helps resolve your issue.


    Regards,
    @Luis_at_Intel

This discussion has been closed.