Some samples from tao_others doesn’t work on Nvidia Jetson AGX Orin 32 GB (dev kit) after firmware update to 4 GB version emulation mode (Jetson Orin Nano 4 GB).
Steps to reproduce:
launch apps/tao_detection samples
for example, ./apps/tao_detection/ds-tao-detection -c configs/TrafficCamNet/traffic_cam_config.txt -i file:////home/dev/projects/deepstream_tao_apps/1.MOV
It doesn’t work by default but launches after replacement to codec x264enc
enc = gst_element_factory_make (“nvv4l2h264enc”, “h264-enc”);
enc = gst_element_factory_make (“x264enc”, “h264-enc”);
replace codec to x264 and launch sample from tao_other - get an error
./deepstream-faciallandmark-app 1 …/…/…/configs/facial_tao/sample_faciallandmarks_config.txt file:////home/dev/projects/deepstream_tao_apps/gaze.mp4 …/…/…/faciallandmark
dev@dev-desktop:~/projects/deepstream_tao_apps/apps/tao_others/deepstream-faciallandmark-app$ ./deepstream-faciallandmark-app 1 …/…/…/configs/facial_tao/sample_faciallandmarks_config.txt file:////home/dev/projects/deepstream_tao_apps/gaze.mp4 …/…/…/faciallandmark
Request sink_0 pad from streammux
OSD and sink elements link failure.
Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU)
• DeepStream Version
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs)
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)
• Hardware Platform (Jetson / GPU) - Jetson AGX Orin as Nano 4GB (flashed AGX Orin Developer Kit)
• DeepStream Version - deepstream-6.2_6.2.0-1_arm64
• JetPack Version (valid for Jetson only) - Version: 5.1.1-b52
• TensorRT Version - 8.5.2-1+cuda11.4
• Issue Type( questions, new requirements, bugs)
Freezing, see the print screens below
• How to reproduce the issue ? (This is for bugs. Including which sample app is using, the configuration files content, the command line used and other details for reproducing)
Download project GitHub - NVIDIA-AI-IOT/deepstream_tao_apps: Sample apps to demonstrate how to deploy models trained with TAO on DeepStream
run ./apps/tao_detection/ds-tao-detection -c configs/yolov4_tao/pgie_yolov4_tao_config.txt -i file:////home/dev/projects/deepstream_tao_apps/1.MOV
To pass issue with hardware encoding/decoding
We have change in this file /apps/tao_detection/deepstream_det_app.c
‘enc = gst_element_factory_make (“nvv4l2h264enc”, "h264-enc»);’
to ‘enc = gst_element_factory_make (“x264enc”, “h264-enc”);’
why do you need to change nvv4l2h264enc to x264enc？ x264enc can only accept software buffer, please add nvvideoconvert before x264enc , please refer to this pipeline:
gst-launch-1.0 filesrc location=/opt/nvidia/deepstream/deepstream/samples/streams/sample_720p.mp4 ! qtdemux ! h264parse ! nvv4l2decoder ! nvvideoconvert ! ‘video/x-raw, format=I420’ ! x264enc bitrate=1000000 ! filesink location=test.264
We want to add some clarification:
Extra 1 - We are aware the Orin nano doesn’t have hardware encoding. When we run our code on the emulated 4gb nano we get the freezing issue, but when we run exactly the same code on the emulated 8gb nano we do not get the freezing issue. What do you think could be the difference when running on 4 vs 8gb emulated? Is it that our current configuration is wrong, but this only causes a problem on 8gb for some reason?
Extra 2 - We find the 4gb works fine with models like dashcamnet from nvidia link but so far does not work with any from github link
There is no update from you for a period, assuming this is not an issue anymore. Hence we are closing this topic. If need further support, please open a new one. Thanks
all Orin Nano have no hardware encoder, please refer to jetson-orin, which sample are you testing? you need to use x264enc to encode. about the freezing issue, please share the code diff and the whole log.
