I would like to ask deepstream pipline questions:

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

i have few questions about deestream pipline:
1, my target is recognite one person with some other objects which like tools, or gun, knife, etc. can i use secondary or more inference? i should how to train? train all objects with person together, or train object one by one to different engine?
2, which deepstream-test is good for me?
3, pls give an advice to push ROI informations to my cloud? which gst plugin should i check?

thanks!

3

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)

4

hello,
iam using jetson NX with deepstream-5.1 and jackpack 4.4.1 or 4.5.1…
and model is yolov4 trained. i trained 4 objects totally together by 4 classes, and main object is person, other 3 objects are tools with person, i want to know what tools on this person’s body, pls give an advice how should i do for the target? i checked deepstream test for car, it can work like that by secondary reference to achieve that task. i may use similar test for me?

thanks

5

I think you can refer https://github.com/NVIDIA-AI-IOT/deepstream_reference_apps/tree/master/back-to-back-detectors

https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_plugin_gst-nvmsgbroker.html

6

thank you i try back-to-back apps now,
another quesitons, if i use nvmsgbroker, should i install kafka at both of could and jetson nx?

thanks

7

hi boss,
i run back-to-back app successfully,
but for secondry_detector_config.txt, i didn’t find where is below fd_lpd file:
model-engine-file=Secondary_FaceDetect/fd_lpd.caffemodel_b1_gpu0_fp32.engine
int8-calib-file=Primary_Detector/cal_trt.bin
i want to use them directly, pls give instructions asap!

thank you

8

Hey, please create a new topic for your new question

9

hi sir,
i got that files already, it is created automatically after run.
thank you very much!