Accessing to first detection model from second detection model

[m.haydari]

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) t4
• DeepStream Version 7.0
• JetPack Version (valid for Jetson only)
• TensorRT Version deepstream 7.0 installation guide
• NVIDIA GPU Driver Version (valid for GPU only) 535
• 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)

hi my pipeline is like this
nvurisecbin → nvstreammux → nvinfer(object detection) → nvinfer(object detection) → nvinfer(classifier) → …
how can i find which object in second nvinfer belong to which object in first nvifner? i know i can access to secondary model from the third one because it’s classifier but there was nothing about this secion which i wrote.

if we can’t access to it pls give me a direct link to where we inference second model by using bounding box of first model to add my own metadata in deepstream source code

for the sake of clarity How can I map each secondary object detection inference to the primary detection that triggered it?

[junshengy]

Please refer to the deepstream-lpr-app. osd_sink_pad_buffer_probe in deepstream_tao_apps/apps/tao_others/deepstream_lpr_app/deepstream-lpr-app/deepstream_lpr_app.c at master · NVIDIA-AI-IOT/deepstream_tao_apps · GitHub

This sample is same as your.

PGIE(car detection) -> SGIE(car license plate detection) -> SGIE(car license plate recognization)

[m.haydari]

my pipeline is like yours but my parent in secondary is always None

[junshengy]

Where in the pipeline do you add the probe function? If the metadata is copied, some information will be lost. Please add the probe function in the src pad of sgie.

[m.haydari]

which part of the deepstream source code can i modify to prevent data loss because i wanna use it after tee

[junshengy]

This part of the code is not open source, you can try adding tee to the downstream of sgie src pad