DeepStream for MTCNN and FaceNet

Accelerated Computing Intelligent Video Analytics DeepStream SDK
1

Hi guys,
I’m working on the MTCNN and FaceNet to recognize the Face, I run on the laptop with 10-11FPS but on Jetson Nano has only 2FPS. How can I use deep stream for this problem??
I want to keep my code, and only integrate deep stream with my code. Thank you so much for helping me. I’m a newbie in using Jetson Nano

3

Hi,

May I know the library you used for inference?
Do you run it with TensorRT, TensorFlow or other frameworks?

Thanks.

4

I run with TensorFlow

5

Hi,

It’s recommended to convert your model into TensorRT for acceleration.
You can find a MTCNN sample in the below GitHub:

Thanks.