can we deploy it in deepstream? if yes, how to deploy it?

[shuxuzhipeng]

[ishang3]

I am interested in the same problem, were you able to see any helpful resources ?

[shuxuzhipeng]

I am interested in the same problem, were you able to see any helpful resources ?

1 hyperpose it also use tensorrt to do the inference and its model file is the format : onnx file. So that it can be deployed in deepstrem. 2 CMU-Perceptual-Computing-Lab/openpose its model file is the format, caffemode which also could be used in deepstream. But u should define the post process customer function to parser the result of inference. Hope u good luck

[Y07yoyo]

Hi @shuxuzhipeng,

Are you able to provide a bit more information on how to use the openpose caffemodel into deepstream?

Kind Regards,

[ishang3]

@Y07yoyo I am solving the same issue - were you successful?

[thancaocuong]

@shuxuzhipeng following this example with custorm yolo, we can write the poseplugin to integrate with deepstream.

[zhink]

deepstream example in here: https://github.com/NVIDIA-AI-IOT/deepstream_pose_estimation