Learning Video Representations using Contrastive Bidirectional Transformer

Chen Sun, Fabien Baradel, Kevin Murphy +1

This paper proposes a self-supervised learning approach for video features that results in significantly improved performance on downstream tasks (such as video classification, cap…