1 paper
Yutao Tang, Benjamin Bejar, Rene Vidal
Recent work on action recognition leverages 3D features and textual information to achieve state-of-the-art performance. However, most of the current few-shot action recognition me…