Delay in streams when we used VPI remap function for dewarping

Autonomous Machines Jetson & Embedded Systems Jetson AGX Xavier
,
1

Dewarping 6 MIPI cameras and ingesting them into RTSP streams on AGX xavier - Our current pipeline works well for 3 cameras, but the moment we add more cameras we see delays in transmission. We have referred to the documentation on RTSP streams and VPI based dewarping for the same.

Gstreamer pipeline :
gst-rtsp-server/examples/test-launch “nvv4l2camerasrc device=/dev/video0 ! video/x-raw(memory:NVMM),format=(string)UYVY, width=(int)1920, height=(int)1080 ! nvvidconv ! nvivafilter customer-lib-name=dewarping.so cuda-process=true ! video/x-raw(memory:NVMM), format=NV12 ! nvvidconv ! nvv4l2h264enc maxperf-enable=1 idrinterval=30 disable-cabac=true insert-aud=true insert-sps-pps=true poc-type=2 bitrate=8000000 ! h264parse ! rtph264pay name=pay0 pt=96” -p 8554

Used this for remapping function : VPI - Vision Programming Interface: Remap

3

Hi,
Please apply this to run all engines at maximum clock:
VPI - Vision Programming Interface: Performance Benchmark

And check if there is improvement.

5

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.