How can I retrain a Faster R-CNN with my own data

I had trained a ssd-mobilenet with my own data (fotos, labels, etc) using Jetson Inference Hello Guide. But I need do the same with a Faster R-CNN model, because ssd-mobilenet istn good for small objects. I need a guide for retrain a Faster R-CNN model, like this jetson-inference/docs/pytorch-ssd.md at master · dusty-nv/jetson-inference · GitHub.

Thnks

Hi,

Unfortunately, we don’t have step-by-step example for the Faster R-CNN.
But you can retrain the model with PyTorch/TorchVision directly.

After training, you can find the below examples for how to convert it into TensorRT for deployment:

Thanks.

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