Learning efficient text-to-image synthesis via interstage cross-sample similarity distillation

doi:10.1007/s11432-020-2900-x

	Learning efficient text-to-image synthesis via interstage cross-sample similarity distillation
	Mao, Fengling 1,2; Ma, Bingpeng 3; Chang, Hong 2,3; Shan, Shiguang 2,3,4; Chen, Xilin 2,3
	2021
发表期刊	SCIENCE CHINA-INFORMATION SCIENCES
ISSN	1674-733X
卷号	64 期号:2 页码:12
摘要	For a given text, previous text-to-image synthesis methods commonly utilize a multistage generation model to produce images with high resolution in a coarse-to-fine manner. However, these methods ignore the interaction among stages, and they do not constrain the consistent cross-sample relations of images generated in different stages. These deficiencies result in inefficient generation and discrimination. In this study, we propose an interstage cross-sample similarity distillation model based on a generative adversarial network (GAN) for learning efficient text-to-image synthesis. To strengthen the interaction among stages, we achieve interstage knowledge distillation from the refined stage to the coarse stages with novel interstage cross-sample similarity distillation blocks. To enhance the constraint on the cross-sample relations of the images generated at different stages, we conduct cross-sample similarity distillation among the stages. Extensive experiments on the Oxford-102 and Caltech-UCSD Birds-200-2011 (CUB) datasets show that our model generates visually pleasing images and achieves quantitatively comparable performance with state-of-the-art methods.
关键词	generative adversarial network (GAN) text-to-image synthesis knowledge distillation
DOI	10.1007/s11432-020-2900-x
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[61876171] ; National Natural Science Foundation of China[61976203] ; Fundamental Research Funds for the Central Universities
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Information Systems ; Engineering, Electrical & Electronic
WOS记录号	WOS:000595379700001
出版者	SCIENCE PRESS
引用统计	被引频次：9[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/2XEOYT63/15960
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Ma, Bingpeng
作者单位	1.ShanghaiTech Univ, Sch Informat Sci & Technol, Shanghai 201210, Peoples R China 2.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Chinese Acad Sci CAS, Beijing 100190, Peoples R China 3.Univ Chinese Acad Sci, Beijing 100049, Peoples R China 4.CAS Ctr Excellence Brain Sci & Intelligence Techn, Shanghai 200031, Peoples R China
推荐引用方式 GB/T 7714	Mao, Fengling,Ma, Bingpeng,Chang, Hong,et al. Learning efficient text-to-image synthesis via interstage cross-sample similarity distillation[J]. SCIENCE CHINA-INFORMATION SCIENCES,2021,64(2):12.
APA	Mao, Fengling,Ma, Bingpeng,Chang, Hong,Shan, Shiguang,&Chen, Xilin.(2021).Learning efficient text-to-image synthesis via interstage cross-sample similarity distillation.SCIENCE CHINA-INFORMATION SCIENCES,64(2),12.
MLA	Mao, Fengling,et al."Learning efficient text-to-image synthesis via interstage cross-sample similarity distillation".SCIENCE CHINA-INFORMATION SCIENCES 64.2(2021):12.