Hello,
I am usig the BD evaluation board for imx8mp:
https://boundarydevices.com/product/universal-smarc-carrier-board/
I am testing object detection capability following the following manual:
https://www.nxp.com/docs/en/user-guide/IMX-MACHINE-LEARNING-UG.pdf page 55
I am trying to modify the following pipeline to work with udp instead of display:
gst-launch-1.0 --no-position v4l2src device=/dev/video3 ! \
video/x-raw,width=640,height=480,framerate=30/1 ! \
tee name=t t. ! queue max-size-buffers=2 leaky=2 ! \
imxvideoconvert_g2d ! \
video/x-raw,width=300,height=300,format=RGBA ! \
videoconvert ! video/x-raw,format=RGB ! \
tensor_converter ! \
tensor_filter framework=tensorflow-lite model=${MODEL} \
custom=Delegate:External,ExtDelegateLib:libvx_delegate.so ! \
tensor_decoder mode=bounding_boxes option1=mobilenet-ssd-postprocess option2=${LABELS} \
option3=0:1:2:3,50 option4=640:480 option5=300:300 ! \
mix. t. ! queue max-size-buffers=2 ! \
imxcompositor_g2d name=mix latency=30000000 min-upstream-latency=30000000
sink_0::zorder=2 sink_1::zorder=1 ! waylandsinkI cannot run the above because my display is not configured and is not needed.
I have tested streaming out using the following pipeline sucessfully with no issues:
$GSTL -v v4l2src device=$USB_CAM_1 ! imxvideoconvert_g2d ! "video/x-raw, width=1920, height=1080, framerate=30/1" ! vpuenc_h264 ! rtph264pay pt=96 ! rtpstreampay ! udpsink host=$HOST_IP port=$HOST_PORT_1 sync=falseWhen I modify desired object detection pipeline to stream out (as shown below), it seems to be running with no errors. However, I do not get any frames (with or without overlay) on the client side.
I have used wireshark to monitor the incoming port on the PC and do not see any frames coming in. I think the issue is imxcompositor_g2d to imxvideoconvert_g2d/vpuenc_h264?
Any suggestions or help is much appreciated.
Thank you.
# DOES NOT ERROR BUT CANNOT SEE THE STREAM
$GSTL -vvv v4l2src device=$USB_CAM_1 ! video/x-raw, width=640, height=480, framerate=30/1 \
! tee name=t t. ! queue leaky=2 max-size-buffers=2 ! imxvideoconvert_g2d ! video/x-raw,width=300,height=300,format=RGBA ! \
videoconvert ! video/x-raw,format=RGB ! \
tensor_converter ! \
tensor_filter framework=tensorflow-lite model=${MODEL} \
custom=Delegate:External,ExtDelegateLib:libvx_delegate.so ! \
tensor_decoder mode=bounding_boxes option1=mobilenet-ssd-postprocess option2=${LABELS} \
option3=0:1:2:3,50 option4=640:480 option5=300:300 ! \
mix. t. ! queue max-size-buffers=2 ! \
imxcompositor_g2d name=mix latency=30000000 min-upstream-latency=30000000 \
sink_0::zorder=2 sink_1::zorder=1 ! \
imxvideoconvert_g2d ! "video/x-raw, width=640, height=480, framerate=30/1,format=RGB16" \
! vpuenc_h264 ! rtph264pay config-interval=1 pt=96 ! rtpstreampay ! udpsink host=$HOST_IP_1 port=$HOST_PORT_1 async=false sync=false
Output of the pipeline: Note that the numbers of incrementing which makes me believe that the pipeline is running. However, on the client side no packets are received.
