Parsing custom network output

[ta-10]

Hello, I have a neural net that takes an input image and produces a depth map.
According to this topic nvinfer supports a custom neural network, and according to the documentation here the plugin can attach raw output tensor data as metadata of type NvDsInferTensorMeta.

Still though I have trouble figuring out how to access the actual output of my network, which would be the depth map. Is there a concrete example that could help?

Setup:
Jetson AGX Xavier
Deepstream 5.0

[bcao]

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)

[bcao]

You can refer deepstream-infer-tensor-meta-test sample

[ta-10]

Thank you for your reply. I managed to get a good grasp at the structure of NvDsInferTensorMeta in order to parse the custom network’s output and found this pose estimation example more helpful to understand how to do so. Cheers!

[bcao]

Great work.