Custom input data to nvinfer

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

I have a pipeline like below,

v4l2src ! videoconvert ! nvvideoconvert ! nvinfer ! nvinfer ! nveglglessink

The overview is that i need 2 inferences to be performed. One is from the frames given from the camera and the other is from the output given from the first nvinfer plugin.
I have read that placing the two nvinfer on after the other takes the input as the output frame from the camera.
How can I pass the inputs of the first nvinfer plugin as the inputs of the second nvinfer plugin?

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

Jetson nano 4gb dev kit
Jetpack 4.5.1 rev1
Deepstream 5.1
TensorRt 7.1.3
Issue type: Question

5

any update on this @kesong ?

6

Do you mean PGIE /SGIE?

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

7

yes

8

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

Here is the samples.

9

I have taken look at the python example and I didn’t find any thing related, can you please point me to the example specifically.

10

operate-on-gie-id this property solved the problem.
https://docs.nvidia.com/metropolis/deepstream/dev-guide/text/DS_plugin_gst-nvinfer.html

12

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