Regarding doubts about deepstream 6.4 tracker implementation

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)-Jetson Orin
• DeepStream Version - 6.4
• JetPack Version (valid for Jetson only) - 6.0 -b52
• TensorRT Version - 8.6.2
• NVIDIA GPU Driver Version (valid for GPU only)-12.4
• Issue Type( questions, new requirements, bugs)- Regarding doubts about deep stream 6.4 tracker and now we are implementing person detection wtih tracking using Yolov8 and deepstream 6.4 tracker when using these two techniques but the tracker id is changing and if the person hide behind the wall then then tracker id should not be changed

how to detect person constantly wherever they can move or in same place

and tracker id should be there same

can you help me to resolve the problem?

• 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)

3

Can you have a try PeopleNet v2.6.1 with NvDCF_accuracy tracker:Gst-nvtracker — DeepStream documentation 6.4 documentation ?

4

ok

will it work more accurately

5

Yes, here is the blog: State-of-the-Art Real-time Multi-Object Trackers with NVIDIA DeepStream SDK 6.2 | NVIDIA Technical Blog

6

Can you have a try PeopleNet v2.6.1 with NvDCF_accuracy tracker:Gst-nvtracker — DeepStream documentation 6.4 documentation ?

can we use yolov8 instead of PeopleNet v2.6.1?

7

Yes, you can use yolov8. But we test it with PeopleNet v2.6.1 for people detect as nvtracker accuracy also depend on the accuracy of people detection.

8

ok thank you