Vision-to-Language Tasks Based on Attributes and Attention Mechanism | IEEE Journals & Magazine | IEEE Xplore