Hello, everyone. I got the following error when run my python demo. I hope you can help me solve it.
import cv2 as cv
net = cv.dnn.readNet('human-pose-estimation-0003.xml', config='human-pose-estimation-0003.bin')
# Specify target device.
# Read an image.
frame = cv.imread('/home/pi/Downloads/1.jpg')
if frame is None:
raise Exception('Image not found!')
# Prepare input blob and perform an inference.
blob = cv.dnn.blobFromImage(frame, size=(456, 256), ddepth=cv.CV_8U)
out = net.forward()
cv2.error: OpenCV(4.3.0-openvino-2020.3.0) ../opencv/modules/dnn/src/ie_ngraph.cpp:600: error: (-2:Unspecified error) Failed to initialize Inference Engine backend (device = MYRIAD): Unexpected network type in function 'initPlugin'
Thank you for posting on the Intel® communities.
Regarding your inquiries, we would like to inform you that we have a forum for those specific products and questions so we are moving it to the Intel® Distribution of OpenVINO™ Toolkit Forum so it can get answered more quickly: https://community.intel.com/t5/Intel-Distribution-of-OpenVINO/bd-p/distribution-openvino-toolkit
Intel Customer Support Technician
Can you share which specific demo you are trying to run?
Just to confirm, are you using Raspberry Pi OS?
Additionally have you ensured that NCS2 has been connected properly?
Yes, I did run on Raspberry Pi OS, and I run a human pose estimation. I have successfully run openvino and ncs2 in the 2019 version. Now I want to try to run it with a faster version of 2020, so I download the corresponding version of openvino and the model file, and the result is wrong. I downloaded the installed package and model files from https://download.01.org/opencv/2020.
The error you encountered is probably caused by a loose NCS2 connection on the USB port (try disconnecting and reconnecting the NCS2 stick).
If you are trying to use the more recent version of OpenVINO, why not try the latest.
Here you can find OpenVINO 2021.1 instruction to installl OpenVINO on PI.
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.