Can I use some parameters with DeepSORT in DeepStream Tracker

I am trying to use DeepSORT in nvtracker.

I read a document and found some parameters in state estimater module.

  • processNoiseVar4Loc
  • processNoiseVar4Size
  • processNoiseVar4Vel
  • measurementNoiseVar4Detector
  • noiseWeightVar4Loc
  • noiseWeightVar4Vel

And it looks that the document recommends that we should use “noseWeightVar4Loc” and “noiseWeightVar4Vel” for DeepSORT.
I would like to know what these are used for and do we have to use noiseWeightVar4Loc and noiseWeightVar4Vel instead of others (, more over can we use procesNoiseVar4Loc and so on with DeepSORT)?

Thank you.

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)


Thank you for your reply.

This is my device information below.
**• Hardware Platform (Jetson / GPU): jetson
**• DeepStream Version: 6.0
**• JetPack Version (valid for Jetson only): 4.6
**• Issue Type( questions, new requirements, bugs): questions

Thank you.

Here is the detail of “State Estimation”:

Gst-nvtracker — DeepStream 6.1.1 Release documentation (


I appreciate for your support.

I’ve already read this section.
And I felt this section recommend me to use these settings blow instead of other configurations such as processNoiseVar4Loc and so on.

  • measurementNoiseVar4Detector
  • measurementNoiseVar4Tracker

Is there any problem to use ‘processNoiseVar4Loc’, ‘processNoiseVar4Size’, ‘processNoiseVar4Vel’ and ‘measurementNoiseVar4Detector’ with DeepSORT?

The reason why DeepSORT uses different params in Kalman filter than NvDCF is because we wanted to follow the same algo as in the original DeepSORT implementation. But, you are free to use the same Kalman filter params from NvDCF for DeepSORT.