The following pipeline works fine.
gst-launch-1.0 -ev \
nvv4l2camerasrc device=/dev/video0 ! \
"video/x-raw(memory:NVMM), format=(string)UYVY, width=(int)1920, height=(int)1080, interlace-mode=progressive, framerate=(fraction)60/1" ! \
nvvideoconvert ! \
nv3dsink
Target
I want to receive the image in GPU memory in a C++ program before nv3dsink, process the image in some way, return it to the pipeline, and display it in nv3dsink.
Question
If I receive image in CPU memory in a C++ program, I can use appsink and appsrc, but what function should I use to receive image in GPU memory?
I often see nvdsosd in deepstream sample programs, is it possible by using it? It is used for drawing bounding boxes and text, rather than image processing in my shallow understanding of this function.
Please note that I am a beginner with 1 month experience with Gstreamer and 1 week experience with Deepstream.
• Hardware Platform (Jetson / GPU): Jetson Orin AGX 32GB Developer kit
• DeepStream Version: 6.1
• JetPack Version (valid for Jetson only): 5.02 (L4T: 35.1)
• TensorRT Version: 8.4.1
• NVIDIA GPU Driver Version (valid for GPU only) : N/A
• Issue Type( questions, new requirements, bugs) Questions