How to train ActionNet from dusty nv?


I am looking at ActionNet

I have a scene that if someone do certain action, I want the Jetson board sends out a signal.
May I know if there is any link to train the ActionNet?
Do I just need to train the ImageNet and feed the onnx to actionNet?


Hi @AK51, these models for ActionNet came from and I haven’t trained them myself before. I believe that they in turn got the models from or used those training scripts (I haven’t tried those though)

Those models use a ResNet backbone, but they have added modifications from an image classifier in order to incorporate the third temporal dimension, hence simply training an ImageNet/ResNet classifier won’t work.

Alternatively, TAO Toolkit supports training of your own action models and can be deployed with DeepStream:

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