Tf-pose-estimation on Jetson Nano and JetPack 4.3

Does anyone has experience with Jetson Nano running tf-pose-estimation under JetPack 4.3? I’m struggeling with a camera stream of a CSI camera under OpenCV.

I get the warning:
[ WARN:0] global /home/nvidia/host/build_opencv/nv_opencv/modules/videoio/src/cap_gstreamer.cpp (933) open OpenCV | GStreamer warning: Cannot query video position: status=0, value=-1, duration=-1
and afterwards the demo script crashes with:
Traceback (most recent call last):
File “run_jetson_nano.py”, line 51, in
logger.info(‘cam image=%dx%d’ % (image.shape[1], image.shape[0]))
AttributeError: ‘NoneType’ object has no attribute ‘shape’

In many forums I’ve found that it has to be a problem with compatibility of OpenCV and the gstreamer.

Hi,

Could you share the source code of run_jetson_nano.py with us?

It looks like the OpenCV cannot get a camera frame correctly.
A common issue is from the GStreamer pipeline command.

For a CSI camera, the pipeline should look like this:
https://github.com/JetsonHacksNano/CSI-Camera/blob/master/simple_camera.py#L24

Thanks.