• Hardware Platform (Jetson / GPU) GPU
• DeepStream Version 6.2
• Graph- Composer 2.5
• TensorRT Version 126.96.36.199
• NVIDIA GPU Driver Version (valid for GPU only) 525.85.12
• Issue Type( questions, new requirements, bugs) question
I have installed graph-composer, and the sample test applications work as expected. However, I want to use graph-composer to check the Peoplenet model, but when I tried to initialise the model from the infer config, it did not display the Peoplenet model in the list of models.
If any further commands need to be run in order to add the PeopleNet model to the Graph Composer, please advise.
adding a reference photo where the subject isn’t in the standard model list.
Have you used the c/c++ DeepStream before? Do you know how gst-nvinfer integrate models?
I’ve used C/C++ deepstream before, and with deepstream 5.1 I incorporated Peoplenet as pgie and a custom model as sgie.
I’m new to graph-composer currently, though.
Hello @Fiona.Chen ,
I was unsure whether the PeopleNet model had any external instructions from Graph Composer that would allow me to add these models to the default list of models, but I was able to do so.
Now that I’ve defined the config infer file and the Model etlt file, I can add the PeopleNet model.
Please follow the instruction here to prepare the peoplenet model and configuration file. deepstream_reference_apps/deepstream_app_tao_configs at master · NVIDIA-AI-IOT/deepstream_reference_apps (github.com)
After the model and configuration file are ready, you can add the model configuration through video inference extension property settings. Don’t forget check mark the option in the UI if you want to enable it.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.