Adding secondary network with nydsanalytics

[user76744]

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU)- Jetson
• DeepStream Version 5.0 (probably)
• JetPack Version (valid for Jetson only)
• TensorRT Version- 7.1.3.0
• 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)

Hey, I want to add a secondary network in the deepstream-nvdsanalytics file to do some inferencing on the detections. Although I’ve changed the config file and linked sgie to the pipeline, I’m not sure whether any classification is happening as no labels are overlayed apart from the detected class and the tracking Id. Can someone help me to overlay the classified value and can anyone tell me which meta the value is stored so that I can access it for another dependent program?

What I’ve noticed is the color of the detected box sometimes changes from red to blue for a few milliseconds. I don’t know the cause but can it be because of the secondary network?
Thank you.

This is my config file:

[property]
gpu-id=0
net-scale-factor=1
onnx-file=/home/apxnx1/Desktop/tensorrt/sendgb-hG4KrWspI6w/gender_googlenet.onnx
model-engine-file=/home/apxnx1/Desktop/tensorrt/code/gender.engine
force-implicit-batch-dim=1
batch-size=1
network-mode=0
network-type=1
input-object-min-width=224
input-object-min-height=224
process-mode=2
model-color-format=1
gpu-id=0
gie-unique-id=2
operate-on-gie-id=1
operate-on-class-ids=2
is-classifier=1
output-tensor-meta=1
output-blob-names=output_cov/Sigmoid
classifier-async-mode=1
classifier-threshold=0.51

[fanzh]

deepstream-nvdsanalytics is used to test nvdsanalytics, it only include one gie. you can use deepstream-test2 to test pgie+sgie.
you can get classification results in NvDsObjectMeta’s NvDsClassifierMeta, please refer to deepstream-infer-tensor-meta-test, you can see how to access NvDsClassifierMeta.

[user76744]

So is there any way to do analytics and classification together on the detections?

[fanzh]

please refer to deepstream-nvdsanalytics-test’s readme, you can build a new deepstream-app, then add [nvds-analytics] to configure file, like source4_1080p_dec_infer-resnet_tracker_sgie_tiled_display_int8.txt, then pipleline will have pgie, trakcer , analytics, sgie…