Semantic-diversity transfer network for generalized zero-shot learning via inner disagreement based OOD detector

doi:10.1016/j.knosys.2021.107337

CORC > 自动化研究所 > 中国科学院自动化研究所 > 模式识别国家重点实验室 > 机器人视觉团队

	Semantic-diversity transfer network for generalized zero-shot learning via inner disagreement based OOD detector
	Liu, Bo 1,2; Dong, Qiulei 1,3,4; Hu, Zhanyi 1,2
刊名	KNOWLEDGE-BASED SYSTEMS
	2021-10-11
卷号	229 页码:11
关键词	Zero-shot learning Visual-semantic embedding Out-of-distribution detection
ISSN号	0950-7051
DOI	10.1016/j.knosys.2021.107337
通讯作者	Dong, Qiulei(qldong@nlpr.ia.ac.cn)
英文摘要	Zero-shot learning (ZSL) aims to recognize objects from unseen classes, where the key is to transfer knowledge from seen classes to unseen classes by establishing appropriate mappings between visual and semantic features. Currently, the knowledge transfer in many existing works is rather limited due to various factors: (i) the widely used visual features are global ones, and they are not completely consistent with semantic attributes; (ii) only one mapping is learned, which is not able to effectively model diverse visual-semantic relations; (iii) the bias problem in the generalized ZSL (GZSL) could not be effectively handled. In this paper, we propose two techniques to alleviate these limitations. Firstly, we propose a Semantic-diversity transfer Network (SetNet) addressing the first two limitations, where (1) a multiple-attention architecture and a diversity regularizer are proposed to learn multiple local visual features being more consistent with semantic attributes and (2) a projector ensemble which geometrically takes diverse local features as inputs is proposed to diversify visual-semantic relations. Secondly, we propose an inner disagreement based domain detection module (ID3M) for GZSL to alleviate the bias problem, which picks out unseen-class data before class-level classification. Due to the lack of unseen-class data in the training stage, ID3M employs a novel self-contained training scheme and detects out unseen-class data based on a proposed inner disagreement criterion. Experimental results on three public datasets show that the proposed SetNet with the explored ID3M achieves a significant improvement against many state-of-the-art methods. (C) 2021 Elsevier B.V. All rights reserved.
资助项目	National Natural Science Foundation of China (NSFC)[61991423] ; National Natural Science Foundation of China (NSFC)[U1805264] ; Strategic Priority Research Program of Chinese Academy of Sciences[XDB32050100]
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:000694743300003
资助机构	National Natural Science Foundation of China (NSFC) ; Strategic Priority Research Program of Chinese Academy of Sciences
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/46034]
专题	自动化研究所_模式识别国家重点实验室_机器人视觉团队
通讯作者	Dong, Qiulei
作者单位	1.Chinese Acad Sci, Inst Automat, Natl Lab Pattern Recognit, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Sch Future Technol, Beijing 100049, Peoples R China 3.Univ Chinese Acad Sci, Sch Artificial Intelligence, Beijing 100049, Peoples R China 4.Chinese Acad Sci, Ctr Excellence Brain Sci & Intelligence Technol, Shanghai 200031, Peoples R China
推荐引用方式 GB/T 7714	Liu, Bo,Dong, Qiulei,Hu, Zhanyi. Semantic-diversity transfer network for generalized zero-shot learning via inner disagreement based OOD detector[J]. KNOWLEDGE-BASED SYSTEMS,2021,229:11.
APA	Liu, Bo,Dong, Qiulei,&Hu, Zhanyi.(2021).Semantic-diversity transfer network for generalized zero-shot learning via inner disagreement based OOD detector.KNOWLEDGE-BASED SYSTEMS,229,11.
MLA	Liu, Bo,et al."Semantic-diversity transfer network for generalized zero-shot learning via inner disagreement based OOD detector".KNOWLEDGE-BASED SYSTEMS 229(2021):11.