Task-Adaptive Attention for Image Captioning

doi:10.1109/TCSVT.2021.3067449

	Task-Adaptive Attention for Image Captioning
	Yan, Chenggang 1,2; Hao, Yiming 1; Li, Liang 3; Yin, Jian 1; Liu, Anan 4; Mao, Zhendong 5; Chen, Zhenyu 6,7; Gao, Xingyu 8
	2022
发表期刊	IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY
ISSN	1051-8215
卷号	32 期号:1 页码:43-51
摘要	Attention mechanisms are now widely used in image captioning models. However, most attention models only focus on visual features. When generating syntax related words, little visual information is needed. In this case, these attention models could mislead the word generation. In this paper, we propose Task-Adaptive Attention module for image captioning, which can alleviate this misleading problem and learn implicit non-visual clues which can be helpful for the generation of non-visual words. We further introduce a diversity regularization to enhance the expression ability of the Task-Adaptive Attention module. Extensive experiments on the MSCOCO captioning dataset demonstrate that by plugging our Task-Adaptive Attention module into a vanilla Transformer-based image captioning model, performance improvement can be achieved.
关键词	Task analysis Visualization Feature extraction Decoding Computational modeling Adaptation models Feeds Image captioning attention mechanism transformer
DOI	10.1109/TCSVT.2021.3067449
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Program of China[2020YFB1406604] ; National Natural Science Foundation of China[61771457] ; National Natural Science Foundation of China[61732007] ; National Natural Science Foundation of China[61931008] ; National Natural Science Foundation of China[61672497] ; National Natural Science Foundation of China[61971268] ; National Natural Science Foundation of China[61772494] ; National Natural Science Foundation of China[62022083]
WOS研究方向	Engineering
WOS类目	Engineering, Electrical & Electronic
WOS记录号	WOS:000742183600008
出版者	IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
引用统计	被引频次：196[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/18267
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Li, Liang
作者单位	1.Shandong Univ, Sch Mech Elect & Informat Engn, Weihai 264209, Peoples R China 2.Hangzhou Dianzi Univ, Dept Automat, Hangzhou 310018, Peoples R China 3.Chinese Acad Sci, Inst Comp Technol, Beijing 100190, Peoples R China 4.Tianjin Univ, Sch Elect & Informat Engn, Tianjin 300072, Peoples R China 5.Univ Sci & Technol China, Sch Informat Sci & Technol, Hefei 230052, Peoples R China 6.State Grid Corp China, Big Data Ctr, Beijing 100031, Peoples R China 7.China Elect Power Res Inst, Beijing 100192, Peoples R China 8.Chinese Acad Sci, Inst Microelect, Beijing 100029, Peoples R China
推荐引用方式 GB/T 7714	Yan, Chenggang,Hao, Yiming,Li, Liang,et al. Task-Adaptive Attention for Image Captioning[J]. IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,2022,32(1):43-51.
APA	Yan, Chenggang.,Hao, Yiming.,Li, Liang.,Yin, Jian.,Liu, Anan.,...&Gao, Xingyu.(2022).Task-Adaptive Attention for Image Captioning.IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY,32(1),43-51.
MLA	Yan, Chenggang,et al."Task-Adaptive Attention for Image Captioning".IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 32.1(2022):43-51.