C3D

The dataset used in this paper is a collection of videos for various tasks such as video recognition and action recognition.

BibTex: