Run human_pose_estimation_demo use GPU error

[Hung__Edwin]

I run human_pose_estimation_demo use GPU
has some error below 
My version is latest version  "l_openvino_toolkit_p_2019.1.144"

./human_pose_estimation_demo -i /dev/video0 -m /home/asrock/inference_engine_samples_build/intel64/human-pose-estimation-0001/human-pose-estimation-0001.xml -d GPU
InferenceEngine: 
    API version ............ 1.6
    Build .................. custom_releases/2019/R1.1_28dfbfdd28954c4dfd2f94403dd8dfc1f411038b
[ INFO ] Parsing input parameters
[ ERROR ] Cannot find plugin to use :Tried load plugin : clDNNPlugin,  error: Plugin clDNNPlugin cannot be loaded: cannot load plugin: clDNNPlugin from : Cannot load library 'libclDNNPlugin.so': libOpenCL.so.1: cannot open shared object file: No such file or directory, skipping

How should I do??
Thank you!
 

[Michael_C_Intel1]

Hi EdwinH,

Like the other posts, I'll move this one to computer vision forum from the OpenCL* forum to get more visibility from OpenVINO stakeholders.

Can you please add the data requested in this guidance? Does the system have an OpenCL device available?

That error is suggestive of OpenVINO prerequisites not being visible or available to OpenVINO libraries. I recommend walking through the prerequisites guide for OpenVINO. Check to make sure hardware and software prerequisites are met. Also walking through the installation guide for Linux.

-MichaelC

[Shubha_R_Intel]

Dear Hung, Edwin,

I just tried this on OpenVino 2019 R1.1 with both FP16 and FP32 on -d GPU and it works fine. As Michael C suggested, I think your OpenCL drivers are not installed, or maybe you have the wrong version installed.  Also please make sure that you ran setupvars.bat first, and from that shell also run the human_pose_estimation_demo.

Hope it helps,

Thanks,

Shubha