Please provide complete information as applicable to your setup.
• Hardware Platform (Jetson / GPU)
Jetson Xavier
• DeepStream Version
6.0
• JetPack Version (valid for Jetson only)
4.6.3
• TensorRT Version
8.2.1
• 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)
I clone into deepstream/deepstream/sources folder
git clone GitHub - NVIDIA-AI-IOT/deepstream_tao_apps: Sample apps to demonstrate how to deploy models trained with TAO on DeepStream
follow README for downloading models.
sudo make bodypose2D app and gesture app
I had to generate engine files for bodypose2D with tao-converter to get it to work. Used the following command:
sudo ./tao-converter -e /opt/nvidia/deepstream/deepstream/sources/deepstream_tao_apps/models/bodypose2d/model.etlt_b32_gpu0_fp16.engine -p input_1:0,1x288x384x3,1x288x384x3,32x288x384x3 -t fp16 -k nvidia_tlt -m 1 nvidia_tlt /opt/nvidia/deepstream/deepstream/sources/deepstream_tao_apps/models/bodypose2d/model.etlt
Then running bodypose2D app works fine. Was even able to write my own custom app using the basic outline of this app.
Now tried to run the gesture app and I get:
Decodebin child added: source
Decodebin child added: decodebin0
Running...
Decodebin child added: nvjpegdec0
In cb_newpad
###Decodebin pick nvidia decoder plugin.
terminate called after throwing an instance of 'std::runtime_error'
what(): invalid input pafmap dimension.
Aborted (core dumped)
After some troubleshooting, the pafmap dimensions are:
c: 36, h: 48, w:19
and heatmap dims are:
c: 144, h: 192, w:38
The app fails at line 433:
bodypose2Dposepost->execute(output, tempPafmap, tempHeatmap, post_w, post_h);
• Requirement details( This is for new requirement. Including the module name-for which plugin or for which sample application, the function description)