Image captioning with triple-attention and stack parallel LSTM

doi:10.1016/j.neucom.2018.08.069

	Image captioning with triple-attention and stack parallel LSTM
	Zhu, Xinxin1,2 ; Li, Lixiang 1,2; Liu, Jing3 ; Li, Ziyi 4; Peng, Haipeng 1,2; Niu, Xinxin 1,2
发表期刊	NEUROCOMPUTING
ISSN	0925-2312
	2018-11-30
卷号	319 页码:55-65
通讯作者	Li, Lixiang(li_lixiang2006@163.com)
摘要	Image captioning aims to describe the content of images with a sentence. It is a natural way for people to express their understanding, but a challenging and important task from the view of image understanding. In this paper, we propose two innovations to improve the performance of such a sequence learning problem. First, we give a new attention method named triple attention (TA-LSTM) which can leverage the image context information at every stage of LSTM. Then, we redesign the structure of basic LSTM, in which not only the stacked LSTM but also the paralleled LSTM are adopted, called as PS-LSTM. In this structure, we not only use the stack LSTM but also use the parallel LSTM to achieve the improvement of the performance compared with the normal LSTM. Through this structure, the proposed model can ensemble more parameters on single model and has ensemble ability itself. Through numerical experiments, on the public available MSCOCO dataset, our final TA-PS-LSTM model achieves comparable performance with some state-of-the-art methods. (c) 2018 Elsevier B.V. All rights reserved.
关键词	Image caption Deep learning LSTM CNN Attention
DOI	10.1016/j.neucom.2018.08.069
收录类别	SCI
语种	英语
资助项目	National Key R&D Program of China[2016YFB0800602] ; National Natural Science Foundation of China[61573067] ; National Natural Science Foundation of China[61771071] ; National Key R&D Program of China[2016YFB0800602] ; National Natural Science Foundation of China[61573067] ; National Natural Science Foundation of China[61771071]
项目资助者	National Key R&D Program of China ; National Natural Science Foundation of China
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000446229200006
出版者	ELSEVIER SCIENCE BV
引用统计	被引频次：42[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/28109
专题	紫东太初大模型研究中心_图像与视频分析
通讯作者	Li, Lixiang
作者单位	1.Beijing Univ Posts & Telecommun, State Key Lab Networking & Switching Technol, Informat Secur Ctr, Beijing 100876, Peoples R China 2.Beijing Univ Posts & Telecommun, Natl Engn Lab Disaster Backup & Recovery, Beijing 100876, Peoples R China 3.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing, Peoples R China 4.Beijing Technol & Business Univ, Beijing 100048, Peoples R China
推荐引用方式 GB/T 7714	Zhu, Xinxin,Li, Lixiang,Liu, Jing,et al. Image captioning with triple-attention and stack parallel LSTM[J]. NEUROCOMPUTING,2018,319:55-65.
APA	Zhu, Xinxin,Li, Lixiang,Liu, Jing,Li, Ziyi,Peng, Haipeng,&Niu, Xinxin.(2018).Image captioning with triple-attention and stack parallel LSTM.NEUROCOMPUTING,319,55-65.
MLA	Zhu, Xinxin,et al."Image captioning with triple-attention and stack parallel LSTM".NEUROCOMPUTING 319(2018):55-65.