Hi, I have downloaded pretrained model from the link given(egogesture_resnetl_10_RGB_8.pth). I had converted it to onnx format. I would like to do an inference with the webcam vedio?
How should I prceed?
for converting to onnx I used resnetl.py file. But there the number of classes is 400. Is it the correct architecture to be followed? Can you please explain the complete steps to do the inference with web cam using pretrained model?