
作者 : Dapeng Chen, Hongsheng Li, Tong Xiao (商汤) 1.解决的问题 Video Person Re-id 2.介绍 现有的很多方法...
题目:Deep Visual-Semantic Alignments for Generating Image Descriptions 作者:Andrej Karpathy...
题目:Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering ...