How to use pretrained ResNet 2+1D to do video classification

Hi,

I am working on a demo about video classification. Could anyone share with me example codes of reading a video, doing the transformation, predicting and outputting the label with pretrained torchvision.models.video.r2plus1d_18 ?

Thanks a lot!

1 Like

I found an example in https://github.com/HHTseng/video-classification/blob/master/ResNetCRNN/UCF101_ResNetCRNN.py, that may be helpful.

1 Like

Thanks a lot! I got some inspiration from it.

Hi,

Did you managed to find some example code?
I also need that, please share if you have any :slight_smile:

Thanks