Which way is the best for traffic sign detection on Jetson Nano?

So, I start a project “Traffic Sign Detection on Nvidia Jetson Nano”, and I want to ask which is the best way to train the system: YOLO, Faster RCNN, Pytorch etc?
I try to train with YOLO, and run with DeepStream 6.0 and it is great, but I need to change train file (for example, now, I need to divide traffic sign in more classes, and I have to train again with YOLO, but it takes me a lot of time).
If you could recommend me some ways to resolve my purpose.

Thanks a lot !

Hi,

This looks like more related to Deepstream and Jetson Nano. We are moving this post to the Jetson Nano forum to get better help.

Thank you.

1 Like

Hi,

It’s recommended to try our TAO toolkit.

Thanks.

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.