Convolutional Reconstruction-to-Sequence for Video Captioning | IEEE Journals & Magazine | IEEE Xplore