Create stream with gstremer

Accelerated Computing Intelligent Video Analytics General
1

Hi!
I use ubuntu, python and gstreamer or c++ and gstreamer.
How I can create stream and send images from GPU?
I found this code, but get error: gi.repository.GLib.Error: gst_parse_error: no element “nvvidconv” (1)

    pipeline = None
    bus = None
    message = None

    # initialize GStreamer
    Gst.init(None)

    for i in range(1, 100):
        print("loop =", i, " ")
        # build the pipeline
        pipeline = Gst.parse_launch(
            "videotestsrc ! video/x-raw,width=1920,height=1080 ! nvvidconv ! video/x-raw(memory:NVMM),width=480,height=380,pixel-aspect-ratio=1/1 ! nvoverlaysink overlay-x=0 overlay-y=360 overlay-w=480 overlay-h=380 "
        )

        # start playing
        print("Switch to PLAYING state")
        pipeline.set_state(Gst.State.PLAYING)

        sleep(5)
        print("Send EoS")
        Gst.Element.send_event(pipeline, Gst.Event.new_eos())
        # wait until EOS or error
        bus = pipeline.get_bus()
        msg = bus.timed_pop_filtered(
            Gst.CLOCK_TIME_NONE, Gst.MessageType.EOS)

        # free resources
        print("Switch to NULL state")
        pipeline.set_state(Gst.State.NULL)
        sleep(1)
2

Sorry for the late response, is this still an issue to support? Thanks

4

Hi,
Do you run in docker? The sample should run on clean-installed system directly. Seems like you run in docker and certain libs are not included. So certain plugins are missing.

5

No, I use ubuntu.

6

Hi,
Do you mean you run the command on host PC? Not run on Xavier NX?

7

I have pc with Ubuntu or linux and me need create stream from video card nvidea

8

Hi,
What is the graphics card you are using? GTX or RTX?

9

I sue RTX

16

nvvidconv is for jetson platform.

17

jetson is a lib or it is a OS? If it is os, then I can use virtual OS for striming from nvidia card?

18

You can not. it’s aarch64 platform.

19

But how I can create stream from nvidia card on PC?

20

use nvvideoconvert instead of nvvidconv on dGPU.

21

Iget error: gst_parse_error: no element “nvvideoconvert” (1)

22

You need to install deepstream sdk for this plugin.
you can try docker for ease use.
https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_docker_containers.html#a-docker-container-for-dgpu

24

I’m use’d docker pull nvcr.io/nvidia/deepstream:6.1.1-devel and install nvidia-container-runtime

docker pull nvcr.io/nvidia/deepstream:6.1.1-devel
sudo docker ps -a

return

CONTAINER ID   IMAGE                                   COMMAND                  CREATED          STATUS    PORTS     NAMES
3495bf0f3f6d   nvcr.io/nvidia/deepstream:6.1.1-devel   "--name nvidia_deeps…"   20 minutes ago   Created             nvidia_deepstream_devel

But
sudo docker start nvidia_deepstream_devel
return error:

Error response from daemon: failed to create shim: OCI runtime create failed: container_linux.go:380: starting container process caused: exec: "--name": executable file not found in $PATH: unknown
Error: failed to start containers: nvidia_deepstream_devel

Whot me need do? I use ubuntu 22.04 or windows 10.

25

docker run --rm -it --gpus device=0 --runtime=nvidia -v /tmp/.X11-unix:/tmp/.X11-unix -e DISPLAY=$DISPLAY IMAGE ID

26

It’s

sudo docker run --rm -it --gpus device=0 --runtime=nvidia -v /tmp/.X11-unix:/tmp/.X11-unix -e DISPLAY=3495bf0f3f6d

return

"docker run" requires at least 1 argument.
See 'docker run --help'.

Usage:  docker run [OPTIONS] IMAGE [COMMAND] [ARG...]

Run a command in a new container

How I can run this container?

27

You can get image id by docker images
REPOSITORY TAG IMAGE ID CREATED SIZE
nvcr.io/nvidia/deepstream 6.1.1-triton ec8cd4081eca 13 days ago 14.4GB

DISPLAY is one environment variable you set by export DISPLAY=**, if you do not need this, you can remove it -e DISPLAY=$DISPLAY

28

This

sudo docker run --rm -it --gpus device=0 --runtime=nvidia -v /tmp/.X11-unix:/tmp/.X11-unix -e DISPLAY=**

and this

sudo docker run --rm -it --gpus device=0 --runtime=nvidia -v /tmp/.X11-unix:/tmp/.X11-unix -e DISPLAY=**

too not work.