Is there a way to ignore some class for tracking when using deepstream-test5?

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) Jetson
• DeepStream Version 6.1.1
• JetPack Version (valid for Jetson only)
• TensorRT Version
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs) questions
• 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)


I want to track people only.
But when using test5, person/vehicle/… will be detected.
And even I changed to use tao_peoplenet, there will still have person/bag/face.
And the tracking result is not good, because bag and face is not stable at all.

Is there a way to filter unneeded classes before tracking?
Thank you.

Is it fine to filter classes in nvinfer?

Hi, @kesong

Thanks for the reply.
But I didn’t find nvinfer either in the config file or deepstream_test5_app_main.c.
How to set this filter?

Can you have a try with “filter-out-class-ids” in nvinfer: Gst-nvinfer — DeepStream 6.1.1 Release documentation

Hi, @kesong

Thanks for your reply.

I tried to add filter-out-class-ids=1;2 to config file of pgie,
and now tracking only has person.

1 Like

Glad to know it works for you.