[Jetson Nano] Running deepsteam demo 'failed to create osd_conv'

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Environment:

jetson-nano-sd-card-image-r32.3.1
deepstream_sdk_v4.0.1_jetson.tbz2

install DS SDK follow this page:

https://docs.nvidia.com/metropolis/deepstream/dev-guide/index.html

when I run the command

cd ./deepstream_sdk_v4.0.1_jetson/samples/configs/deepstream-app
deepstream-app -c source8_1080p_dec_infer-resnet_tracker_tiled_display_fp16_nano.txt

got fail info

** ERROR: <create_osd_bin:40>: Failed to create 'osd_conv'
** ERROR: <create_osd_bin:114>: create_osd_bin failed
** ERROR: <create_processing_instance:705>: create_processing_instance failed
** ERROR: <create_pipeline:1045>: create_pipeline failed
** ERROR: <main:632>: Failed to create pipeline
Quitting
App run failed

and then i disable the plugin ‘osd’, run again

Unable to set device in gst_nvstreammux_change_state
Unable to set device in gst_nvstreammux_change_state
Unable to set device in gst_nvstreammux_change_state
** ERROR: <main:651>: Failed to set pipeline to PAUSED
Quitting
Unable to set device in gst_nvstreammux_change_state
App run failed

finally i disable the plugin ‘streammux’

the the video runs without track abilities.

How to fix this problem? Appreciate it!

2

yep, change version to 4.0.2 and fix it.

Dependence not match, even I have immigrate the libnvinfer.so.5 from jetpack 4.2.1.

And I check the platform compatability, maybe is the openCV version not match.

Anyway, I fixed it.