Is Space-Time Attention All You Need for Video Understanding?

                 Gedas Bertasius 1 Heng Wang 1 Lorenzo Torresani 1 2

             Abstract               Video understanding shares several high-level similarities
                                with NLP. First of all, videos and sentences are both sequen-
    We present a convolution-free approach to video
                                tial. Furthermore, precisely as the meaning of a word can
    classification built exclusively ...