Vision-Enhanced and Consensus-Aware Transformer for Image Captioning | IEEE Journals & Magazine | IEEE Xplore