Intel® Distribution of OpenVINO™ Toolkit
Community support and discussions about the Intel® Distribution of OpenVINO™ toolkit, OpenCV, and all things computer vision-related on Intel® platforms.
5922 Discussions

Failed to run "End to end video analytics end2end_video_analytics_ie Demo" on Visual Studio 2015


Hello, I've been trying to get running under Windows 10 under Visual Studio 10. I managed to get it to compile and run. But I found out it's failing to open "cars_768x768.h264". Is the opencv under openvino folder sufficient enought to run the demo? Is there anywhere I can find steps to rebuilt opencv under the openvino folder?

Thank you.


P.S I just realized I might have double-posted. I have no idea how to delete it since I can't see it. Sorry for the trouble.

0 Kudos
5 Replies


Dear zero, heisenburgzero

I believe i have answered this question in the above-referenced forum post.

Thanks for using OpenVino !




I tried to run the sample application end2end video analytics.

I got the following  error.Can anyone help me to solve this error

root@ioz-NUC8i3BEH:/home/ioz/desktop/samples/build/intel64/Release# ./end2end_video_analytics_ie -i=desktop/samples/end2end_video_analytics/test_content/video/cars_768x768.h264 -d CPU -m=Desktop/IR/FP32/mobilenet-ssd.xml -l=desktop/samples/end2end_video_analytics/test_content/IR/SSD/pascal_voc_classes.txt

    == Inference Engine Plugin Information ==
    Plugin version ......... 1.6
    Plugin name ............ MKLDNNPlugin
    Plugin build ........... 23780

Failed to load network:

Thanks in advance



Dear G, Shanmuga vadivelu,

Are you on 2019R1.1 ? I have no problem running end2end_video_analytics_ie.exe on Windows 10. (OS should not matter).  Please follow the 

end to end demo doc

end to end ubuntu instructions

One of the steps is to use to download the required model. And instructions are given in the first doc above regarding how to convert the model into Model Optimizer Generated IR.

If you follow these instructions carefully and you are using OpenVino 2019R1.1 you should meet with success. Below are my results:

C:\Users\sdramani\Documents\Intel\OpenVINO\inference_engine_samples_build\intel64\Release>python "c:\Program Files (x86)\IntelSWTools\openvino_2019.1.148\deployment_tools\model_optimizer\" --input_model "c:\Program Files (x86)\IntelSWTools\openvino_2019.1.148\deployment_tools\tools\model_downloader\object_detection\common\mobilenet-ssd\caffe\mobilenet-ssd.caffemodel" --scale 255.0 --mean_values [123.68,116.779,103.939]  --data_type FP32
Model Optimizer arguments:
Common parameters:
        - Path to the Input Model:      c:\Program Files (x86)\IntelSWTools\openvino_2019.1.148\deployment_tools\tools\model_downloader\object_detection\common\mobilenet-ssd\caffe\mobilenet-ssd.caffemodel
        - Path for generated IR:        C:\Users\sdramani\Documents\Intel\OpenVINO\inference_engine_samples_build\intel64\Release\.
        - IR output name:       mobilenet-ssd
        - Log level:    ERROR
        - Batch:        Not specified, inherited from the model
        - Input layers:         Not specified, inherited from the model
        - Output layers:        Not specified, inherited from the model
        - Input shapes:         Not specified, inherited from the model
        - Mean values:  [123.68,116.779,103.939]
        - Scale values:         Not specified
        - Scale factor:         255.0
        - Precision of IR:      FP32
        - Enable fusing:        True
        - Enable grouped convolutions fusing:   True
        - Move mean values to preprocess section:       False
        - Reverse input channels:       False
Caffe specific parameters:
        - Enable resnet optimization:   True
        - Path to the Input prototxt:   c:\Program Files (x86)\IntelSWTools\openvino_2019.1.148\deployment_tools\tools\model_downloader\object_detection\common\mobilenet-ssd\caffe\mobilenet-ssd.prototxt
        - Path to CustomLayersMapping.xml:      c:\Program Files (x86)\IntelSWTools\openvino_2019.1.148\deployment_tools\model_optimizer\extensions\front\caffe\CustomLayersMapping.xml
        - Path to a mean file:  Not specified
        - Offsets for a mean file:      Not specified
Model Optimizer version:        2019.1.1-83-g28dfbfd

[ SUCCESS ] Generated IR model.
[ SUCCESS ] XML file: C:\Users\sdramani\Documents\Intel\OpenVINO\inference_engine_samples_build\intel64\Release\.\mobilenet-ssd.xml
[ SUCCESS ] BIN file: C:\Users\sdramani\Documents\Intel\OpenVINO\inference_engine_samples_build\intel64\Release\.\mobilenet-ssd.bin
[ SUCCESS ] Total execution time: 6.82 seconds.

C:\Users\sdramani\Documents\Intel\OpenVINO\inference_engine_samples_build\intel64\Release>end2end_video_analytics_ie.exe -i "c:\Program Files (x86)\IntelSWTools\openvino_2019.1.148\inference_engine\samples\end2end_video_analytics\test_content\image\cars_1920x1080.jpg" -l  "c:\Program Files (x86)\IntelSWTools\openvino_2019.1.148\inference_engine\samples\end2end_video_analytics\test_content\IR\SSD\pascal_voc_classes.txt" -m mobilenet-ssd.xml

        == Inference Engine Plugin Information ==
        Plugin version ......... 1.6
        Plugin name ............ MKLDNNPlugin
        Plugin build ........... 23780

Network loaded.
Batch size = 1
Setting-up input, output blobs...
inputDims=300 300 3 1
outputDims=1 1 100 7
Loading model to plugin...
Create infer request...
plane           not used
bicycle         not used
bird            not used
boat            not used
bottle          not used
bus             not used
car             used
cat             not used
chair           not used
cow             not used
table           not used
dog             not used
horse           not used
motorcy         not used
person          used
plant           not used
sheep           not used
sofa            not used
train           not used
monitor         not used
Running inference...

Batch: 1/256
        pre-stage:      4.63 ms/frame
        infer:          20.79 ms/frame
        post-stage:     5.80 ms/frame

> Pre-stage average:    4.63 ms/frame (decoding, color converting, resizing)
> Infer average:        20.79 ms/frame (inferencing)
> Post-stage average:   5.80 ms/frame (drawing bounding box, encoding, saving)

> Total elapsed execution time: 0.04 sec



Thanks !






Thanks for your help Shubha

I got the result as expected. I would like to know, how can I implement OpenVINO in my own custom object detection using Tensorflow?, as I have trained model(Faster R-CNN Inception_v2_COCO)



Shanmuga vadivelu



Dear G, Shanmuga vadivelu,

First you would follow Tensorflow Training Custom Object Detector, then after you've obtained a trained model, you'd convert it using Model Optimizer Tensorflow Object Detection API instructions .

Finally you'd run inference using as a base one of our Object Detection samples, which we offer in both Python and C++.

Hope it helps,