Video2vec: Learning semantic spatio-temporal embeddings for video representation

Video2vec: Learning semantic spatio-temporal embeddings for video representation