I’d like to ingest a rstp video and “leave” in the GPU memory to be ready for tensorRT inference. I have Ubuntu Server 18.04 with Cuda 11.1, Drivers 455, Tensort 7.2.1.1
Cuda compilation tools, release 11.1, V11.1.105
Build cuda_11.1.TC455_06.29190527_0
I setup Gstreamer on Ubuntu Server 18.04 with the following official command (here):
sudo apt-get install libgstreamer1.0-0 gstreamer1.0-plugins-base gstreamer1.0-plugins-good gstreamer1.0-plugins-bad gstreamer1.0-plugins-ugly gstreamer1.0-libav gstreamer1.0-doc gstreamer1.0-tools gstreamer1.0-x gstreamer1.0-alsa gstreamer1.0-gl gstreamer1.0-gtk3 gstreamer1.0-qt5 gstreamer1.0-pulseaudio
But the following pipeline (I think the camera should be public)
gst-launch-1.0 rtspsrc location=rtsp://wowzaec2demo.streamlock.net/vod/mp4:BigBuckBunny_115k.mov latency=10 num-buffers=256 ! decodebin ! nvvideoconvert ! video/x-raw\(memory:NVMM\),format=BGRx ! fakesink name=s
gives me the error:
GStreamer warning: Error opening bin: no element "nvvideoconvert"
Also, cv2.getBuildInformation()
reports this:
Video I/O:
DC1394: YES (2.2.5)
FFMPEG: YES
avcodec: YES (57.107.100)
avformat: YES (57.83.100)
avutil: YES (55.78.100)
swscale: YES (4.8.100)
avresample: YES (3.7.0)
GStreamer: YES (1.14.5)
v4l/v4l2: YES (linux/videodev2.h)
Do I need Deepstream? If not, how can I solve this? I don’t think Deepstream support my current cuda version.