I3D models trained on Kinetics Pytorch

this repo implements the network of I3D with Pytorch, pre-trained model weights are converted from tensorflow.

Sample code

you can evaluate sample

python evaluate_sample.py

The sample video can be found in /data. Use optical_flow.py to preprocess data to fed for inference

python optical_flow.py path_to_video

This will create two .npy files namely flow.npy and frames.npy, which can be used by the model to classify the video.

The original labels from the Kinetics dataset are in /data.

Reference:

kinetics-i3d
tensorflow-model-zoo.torch