NvDCF Tracker not working for fast moving objects

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

• Hardware Platform (Jetson / GPU)
Jetson Orin
• DeepStream Version
6.1.1
• JetPack Version (valid for Jetson only)
5.0.2
• TensorRT Version
8.4.1-1+cuda11.4

I noticed some objects were not being detected by my python Deepstream app. My model is not the problem, I already tested the same sample video with a tensorrt only inference and it works. I also removed the nvtracker plugin and I see the airplane detection. It doesn’t work when I add the tracker.

How can I modify the tracker’s config file to detect fast-moving objects that are in the frame just for a few seconds?

2

Setting probationAge: 1 in the tracker’s .yml config file fixed my issue.

4

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