pedestrian_tracker_demo - Tracker.cpp

Gorza__Lorenzo · ‎01-14-2020

Hi,

I'm trying to execute the pedestrian tracker demo, but when I process the tracker in the cpp code, loading the weights and the topology of the network, I get the exception:

Exception: Dynamic batch is not enabled.

anyone has had the same issue?

Thanks.

JesusE_Intel · ‎01-23-2020

Hi Lorenzo,

Thanks for reaching out! I was able to run the pedestrian_tracker_demo without any issues using the following command.

./pedestrian_tracker_demo -i person-bicycle-car-detection.mp4 -m_det intel/person-detection-retail-0013/FP16/person-detection-retail-0013.xml -m_reid intel/person-reidentification-retail-0031/FP16/person-reidentification-retail-0031.xml

Could you share additional information about your setup?

Which OpenVINO toolkit version are you using? Please make sure you are using the latest OpenVINO 2019 R3.1
Are you using Ubuntu, Windows or MacOS?
Which networks are you using?
How did you compile the cpp demo?

Regards,

Jesus

R_P___Jose · ‎04-11-2020

Hi there,

I have "collided" with the same demo, but a different scream.

It is running in a fresh centos 7 headless vmware local test instance installed and compiled strictly following the guide (yes, as root, pardon the risk). But I have to figure out how to avoid using GUI and how to avid the cryptic problems raised:

[root@localhost ~]# /root/omz_demos_build/intel64/Release/pedestrian_tracker_demo -i /root/videos/videoplayback.mp4 -m_det /opt/intel/openvino_2020.1.023/deployment_tools/open_model_zoo/tools/downloader/intel/person-detection-retail-0013/FP32/person-detection-retail-0013.xml -m_reid /root/videos/videoplayback.mp4 -m_det /opt/intel/openvino_2020.1.023/deployment_tools/open_model_zoo/tools/downloader/intel/person-reidentification-retail-0031/FP32/person-reidentification-retail-0031.xml d_det CPU -no_show -out /root/result.log

InferenceEngine: 0x7f128fffd030
Loading device CPU
   CPU
   MKLDNNPlugin version ......... 2.1
   Build ........... 37988

[ ERROR ] Person Detection network output layer(ip_reid) should be DetectionOutput, but was FullyConnected

Any input about where is the grip fo this problem will be greatly appreciated.

Cheers,
Jose
(Madrid, Spain)