smart class room demo not working

[verma__Ashish]

I have build inference engine samples and trying to run smart classroom demo using this command -

./smart_classroom_demo -i cam -m_act /opt/intel/computer_vision_sdk/deployment_tools/intel_models/person-detection-retail-0013/FP32/person-detection-retail-0013.xml -m_fd /opt/intel/computer_vision_sdk/deployment_tools/intel_models/face-detection-retail-0004/FP32/face-detection-retail-0004.xml -m_lm /opt/intel/computer_vision_sdk/deployment_tools/intel_models/facial-landmarks-35-adas-0001/FP32/facial-landmarks-35-adas-0001.xml -m_reid /opt/intel/computer_vision_sdk/deployment_tools/intel_models/face-reidentification-retail-0071/FP32/face-reidentification-retail-0071.xml

but I am getting following error -

InferenceEngine:
    API version ............ 1.4
    Build .................. 17328
[ INFO ] Parsing input parameters
Reading video 'cam'
Loading plugin CPU

    API version ............ 1.4
    Build .................. lnx_20181004
    Description ....... MKLDNNPlugin
[ ERROR ] Dims and format are inconsistent.

Any help will be appreciated

Regards

Ashish

[nikos1]

Hi Ashish,

Seems you may be using the wrong path to IR. Check this https://software.intel.com/en-us/articles/OpenVINO-IE-Samples

Anyway, I would try with less parameters first to narrow down the issue.

For example in the first part of your command 

./smart_classroom_demo -i cam -m_act /opt/intel/computer_vision_sdk/deployment_tools/intel_models/person-detection-retail-0013/FP32/person-detection-retail-0013.xml

it may be that you need an action IR for -m_act , you could try again with 

./smart_classroom_demo -i cam -m_act /opt/intel/computer_vision_sdk/deployment_tools/intel_models/person-detection-action-recognition-0004/FP32/person-detection-action-recognition-0004.xml 

[verma__Ashish]

Hi Nikos,

Thanks for your fast reply.

Now I am able to run this command -

./smart_classroom_demo -i cam -m_act /opt/intel/computer_vision_sdk/deployment_tools/intel_models/person-detection-action-recognition-0003/FP32/person-detection-action-recognition-0003.xml -m_fd /opt/intel/computer_vision_sdk/deployment_tools/intel_models/face-detection-retail-0004/FP32/face-detection-retail-0004.xml -m_lm /opt/intel/computer_vision_sdk/deployment_tools/intel_models/facial-landmarks-35-adas-0001/FP32/facial-landmarks-35-adas-0001.xml

but when i am adding flag for reidentification  -m_reid /opt/intel/computer_vision_sdk/deployment_tools/intel_models/face-reidentification-retail-0071/FP32/face-reidentification-retail-0071.xml

I am getting the same error, without mentioning  this flag its working.

In the Inference Engine samples documentation they are using

face-reidentification-retail-0095 model which is not available in intel models of openvino

Regards

Ashish

[nikos1]

Hi Ashish,

What happens if you try -m /opt/intel/computer_vision_sdk/deployment_tools/intel_models/face-reidentification-retail-0095/FP32/face-reidentification-retail-0095.xml

Please note the version or release numbers may be different. In my case face-reidentification-retail-0071.xml is now face-reidentification-retail-0095.xml

If not try to find it

find /opt/intel -name "*face-reidentification*xml"

> face-reidentification-retail-0095 model which is not available in intel models of openvino

Not sure hwy not, it is here in the R5 SDK :  

/home/user/intel/computer_vision_sdk_2018.5.445/deployment_tools/intel_models/face-reidentification-retail-0095/FP32/face-reidentification-retail-0095.xml
/home/user/intel/computer_vision_sdk_2018.5.445/deployment_tools/intel_models/face-reidentification-retail-0095/FP16/face-reidentification-retail-0095.xml

Cheers,

nikos