Kelly0510/FitAQA
Viewer • Updated • 5.51k • 570 • 3
The DeepMind Kinetics dataset is expanded to 700 action classes with at least 600 video clips per class, accompanied by updated statistics and I3D baseline results.
We describe an extension of the DeepMind Kinetics human action dataset from 600 classes to 700 classes, where for each class there are at least 600 video clips from different YouTube videos. This paper details the changes introduced for this new release of the dataset, and includes a comprehensive set of statistics as well as baseline results using the I3D neural network architecture.
No model linking this paper
No Space linking this paper
No Collection including this paper