Jetson performance accessment for my object detection app (DS6 python) from multi rtsp stream

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)
Jetson NX
• DeepStream Version
DS6
• JetPack Version (valid for Jetson only)
4.6 or 4.6.1
• 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)

There’re mutiple cameras in my production scenario, I’m planning to use Jetson NX or higher spec hardware to deploy a custom DS6 python app to implement the object detection for all of the cameras:

  • All cameras output 720p or 1080p with h.264 encoding, FPS30 streams by RTSP
    so my DS6 python app needs to read the streams by RTSP.
  • My DS6 python app is using a single TAO retrained detectnet_v2 resnet18 model for inference
    so the inferencing is based on single one model.
  • Requires inference perf can achive 30FPS

I read the offical benchmarks, but the cases listed there are all for single stream, I’m not sure how to assess with my case of multiple RTSP stream, Could you help?

3

Performance depends on your actual pipeline and configurations. The best way is to test with the real pipeline.

5

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