Social Image Captioning with Tags-Based Attention Model
2018 IEEE International Conference on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData)(2018)
摘要
Recently, image captioning has attracted much attention in machine learning and computer vision. It is not only an important practical application, but also a challenging work. However, related works normally take traditional attention mechanism to tell the model where to look without considering people's intentions. In this paper, we propose a novel tags-based attention mechanism to capturing the people's intentions to generate a human-based sentence. The proposed method can automatically focus on the specific image region where the people want to focus on. The experiments are conducted on MSCOCO dataset, and the results show that our algorithm outperforms the previous methods.
更多查看译文
关键词
Handheld computers,Internet of Things,Green computing,Social computing,Blockchain,Information technology,Erbium
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络