NVDCF Tracker without DeepStream Pipeline


I am working on a Yolo V4 based object detector application which running on a triton inference server. The current implementation is based on python language and doesn’t use any DeepStream pipeline.

I planning to integrate a tracker in my object detection pipeline, and the NVDCF tracker is one of the choices. I want to know how to integrate the NVDCF tracker to a python object detection code (Note: Only NVDCF tracker, no other DeepStream elements/pipeline). Is that possible? Is there any reference sample application available?

Thanks in advance.

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: NVIDIA T4 GPU
• DeepStream Version: DeepStream 6.0
• TensorRT Version: 8.2.2
• NVIDIA GPU Driver Version: 470.103.01
• Issue Type: Questions, Requirement
• Requirement details: Standalone NVDCF tracker API for object detection application (Currently NVDCF tracker is coupled with DeepStream SDK)

Hi @kayccc,

Any thoughts?


Hi @captain_billgardo ,
I don’t think it works since NvDCF tracker is based on the DeepStream Metadata, without DeepStream pipeline, no DeepStream Metadata, NvDCF will not work

1 Like

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.