Trying to run YOLOV3 using deepstream5 but segmentation fault

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

I’m trying to run yolov3 with deepstream5 following the steps in this link: How to detect objects with Nvidia Deepstream 4.0 and YOLO in 5 minutes - Deep Learning and AI related blog

desktop:/opt/nvidia/deepstream/deepstream-5.0/sources/objectDetector_Yolo$ deepstream-app -c deepstream_app_config_yoloV3.txt
Warn: ‘threshold’ parameter has been deprecated. Use ‘pre-cluster-threshold’ instead.

Using winsys: x11
gstnvtracker: Loading low-level lib at /opt/nvidia/deepstream/deepstream-5.0/lib/libnvds_mot_klt.so
gstnvtracker: Failed to open low-level lib at /opt/nvidia/deepstream/deepstream-5.0/lib/libnvds_mot_klt.so
dlopen error: libvisionworks.so: cannot open shared object file: No such file or directory
gstnvtracker: Failed to initilaize low level lib.
** ERROR: main:651: Failed to set pipeline to PAUSED
Quitting
App run failed
Segmentation fault (core dumped)

• Jetson
• DeepStream5.0
• JetPack Version 4.4
• TensorRT 7.1
• cuda 10.2

3

It appears that deepstream-app was not able to load the library file for the tracker (libnvds_mot_klt.so). Can you check if the library file is present at /opt/nvidia/deepstream/deepstream-5.0/lib/ ? If yes, you can check the permissions on the file. For some reason if you don’t have read or execute permissions your your user you can try changing that or try running the deepstream-app as sudo.