Improving Video Captioning with Temporal Composition of a Visual-Syntactic Embedding | IEEE Conference Publication | IEEE Xplore