How do I run facedetect on deepstream

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

How do I run facedetect model on deepstream in jetson nano. I get a bit mind-boggled on how to approach this

1 Like
3

Hi,

Suppose you are using a customized facedetect model.

1. Convert your model into ONNX format.
You can do this with tf2onnx, or PyTorch2onnx.

2. Change the model path and output configure based on your model in the configure file.

For example:

Thanks.

4

I’m using the facedetect model on NGC, not a custom model, which is also the facenet.tlt.

5

Hi,

Please check below sample to run TLT model with Deepstream:

Thanks.