The goal of the Kinetics dataset is to help the computer vision and machine learning communities advance models for video understanding. Given this large human action classification dataset, it may be possible to learn powerful video representations that transfer to different video tasks.
To approach this task in a practical and helpful manner, let's break down the given string into its components and analyze each part systematically.
To approach this task in a practical and helpful manner, let's break down the given string into its components and analyze each part systematically.
1. Possible to use ImageNet checkpoints?
We allow finetuning from public ImageNet checkpoints for the supervised track -- but a link to the specific checkpoint should be provided with each submission.
2. Possible to use optical flow?
Flow can be used as long as not trained on external datasets, except if they are synthetic.
To approach this task in a practical and
3. Can we train on test data without labels (e.g. transductive)?
No.
To approach this task in a practical and
4. Can we use semantic class label information?
Yes, for the supervised track.
To approach this task in a practical and
5. Will there be special tracks for methods using fewer FLOPs / small models or just RGB vs RGB+Audio in the self-supervised track?
We will ask participants to provide the total number of model parameters and the modalities used and plan to create special mentions for those doing well in each setting, but not specific tracks.