Convolutional prototype learning for zero-shot recognition

doi:10.1016/j.imavis.2020.103924

	Convolutional prototype learning for zero-shot recognition
	Liu, Zhizhe 1,2; Zhang, Xingxing 1,2; Zhu, Zhenfeng 1,2; Zheng, Shuai 1,2; Zhao, Yao 1,2; Cheng, Jian3
发表期刊	IMAGE AND VISION COMPUTING
ISSN	0262-8856
	2020-06-01
卷号	98 页码:8
通讯作者	Zhu, Zhenfeng(zhfzhu@bjtu.edu.cn)
摘要	Zero-shot learning (ZSL) has received increasing attention in recent years especially in areas of fine-grained object recognition, retrieval, and image captioning. The key to ZSL is to transfer knowledge from the seen to the unseen classes via auxiliary class attribute vectors. However, the popularly learned projection functions in previous works cannot generalize well since they assume the distribution consistency between seen and unseen domains at sample-level. Besides, the provided non-visual and unique class attributes can significantly degrade the recognition performance in semantic space. In this paper, we propose a simple yet effective convolutional prototype learning (CPL) framework for zero-shot recognition. By assuming distribution consistency at task-level, our CPL is capable of transferring knowledge smoothly to recognize unseen samples. Furthermore, inside each task, discriminative visual prototypes are learned via a distance based training mechanism. Consequently, we can perform recognition in visual space, instead of semantic space. An extensive group of experiments are then carefully designed and presented, demonstrating that CPL obtains more favorable effectiveness, over currently available alternatives under various settings. (C) 2020 Elsevier B.V. All rights reserved.
关键词	Zero-shot recognition Prototype learning Image recognition Deep learning
DOI	10.1016/j.imavis.2020.103924
收录类别	SCI
语种	英语
资助项目	Science and Technology Innovation 2030 -New Generation Artificial Intelligence Major Project[2018AAA0102101] ; National Natural Science Foundation of China[61976018] ; National Natural Science Foundation of China[61532005] ; Fundamental Research Funds for the Central Universities of China[2018JBZ001] ; Fundamental Research Funds for the Central Universities of China[2019YJS048]
项目资助者	Science and Technology Innovation 2030 -New Generation Artificial Intelligence Major Project ; National Natural Science Foundation of China ; Fundamental Research Funds for the Central Universities of China
WOS研究方向	Computer Science ; Engineering ; Optics
WOS类目	Computer Science, Artificial Intelligence ; Computer Science, Software Engineering ; Computer Science, Theory & Methods ; Engineering, Electrical & Electronic ; Optics
WOS记录号	WOS:000536040700006
出版者	ELSEVIER
引用统计	被引频次：12[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/173211/39535
专题	复杂系统认知与决策实验室_高效智能计算与学习
通讯作者	Zhu, Zhenfeng
作者单位	1.Beijing Jiaotong Univ, Inst Informat Sci, Beijing, Peoples R China 2.Beijing Key Lab Adv Informat Sci & Network Techno, Beijing, Peoples R China 3.Chinese Acad Sci, Inst Automat, NLPR, Beijing, Peoples R China
推荐引用方式 GB/T 7714	Liu, Zhizhe,Zhang, Xingxing,Zhu, Zhenfeng,et al. Convolutional prototype learning for zero-shot recognition[J]. IMAGE AND VISION COMPUTING,2020,98:8.
APA	Liu, Zhizhe,Zhang, Xingxing,Zhu, Zhenfeng,Zheng, Shuai,Zhao, Yao,&Cheng, Jian.(2020).Convolutional prototype learning for zero-shot recognition.IMAGE AND VISION COMPUTING,98,8.
MLA	Liu, Zhizhe,et al."Convolutional prototype learning for zero-shot recognition".IMAGE AND VISION COMPUTING 98(2020):8.