CVPR 2018 Oral：Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering

題目：Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering 作者 : Peter Anderson , Xiaodong He , Chris Buehler , Damien Tency 論文地址[https://arxiv.org/abs/1707.07998] 1.解決的問題 I

>>阅读原文<<